Method and apparatus for predicting CSI in cellular system

Abstract

The present disclosure relates to a 5G or 6G communication system for supporting a higher data transmission rate. Methods and apparatus for predicting channel state information (CSI) feedback in a cellular system are provided. A method for a user equipment (UE) to report CSI includes receiving first information related to monitoring performance of a machine learning (ML) model for predicting CSI, second information related to transmitting a performance monitoring report, third information related to reception of a CSI reference signal (CSI-RS) for determining real CSI on a cell, and a CSI-RS based on the third information. The method further includes determining a true CSI based on the reception of the CSI-RS, and determining a performance monitoring report for the ML model based on the first information and the true CSI. The method also includes transmitting a channel with a performance monitoring report based on the second information.

Description

Technical Field

[0001] The present disclosure relates generally to wireless communication systems, and more particularly, to methods and apparatus for predicting future channel state information (CSI) in a cellular system. Background Art

[0002] 5G mobile communication technology defines a wide frequency band, making high transmission rates and new services possible, and can be realized not only in the "below 6 GHz" frequency band such as 3.5 GHz, but also in the "above 6 GHz" frequency band known as millimeter wave (including 28 GHz and 39 GHz). In addition, the implementation of 6G mobile communication technology (called super 5G system) in the terahertz frequency band (such as the 95 GHz to 3 THz band) has been considered in order to achieve a transmission rate fifty times faster than 5G mobile communication technology and an ultra-low latency of one-tenth of 5G mobile communication technology.

[0003] At the beginning of the development of 5G mobile communication technology, in order to support services and meet the performance requirements related to enhanced mobile broadband (eMBB), ultra-reliable and low-latency communication (URLLC) and massive machine type communication (mMTC), standardization of the following technologies has been ongoing: beamforming and massive multiple-input multiple-output (MIMO) for reducing radio wave path loss and increasing radio wave transmission distance in millimeter waves; support parameter sets for dynamic operation of efficient utilization of millimeter wave resources and time slot formats (for example, operation of multiple subcarrier spacings); initial access technology for supporting multi-beam transmission and wide frequency bands; definition and operation of bandwidth parts (BWPs); new channel coding methods such as low-density parity-check (LDPC) codes for large-scale data transmission and polar codes for highly reliable transmission of control information; L2 preprocessing; and network slicing for providing dedicated networks dedicated to specific services.

[0004] Currently, with respect to services supported by 5G mobile communication technology, the industry is continuously discussing improvements and performance enhancements on initial 5G mobile communication technology, and has completed physical layer standardization on technologies such as: vehicle-to-everything (V2X) for assisting driving decisions based on information about the location and status of a vehicle sent by a vehicle and for improving user convenience, new radio unlicensed (NR-U) for system operations that comply with various regulatory-related requirements in unlicensed frequency bands, new radio (NR) user equipment (UE) power saving, non-terrestrial network (NTN) as UE satellite direct communication for ensuring coverage in areas where communication with terrestrial networks is not possible, and positioning.

[0005] In addition, the standardization of air interface architecture / protocols for technologies such as the Industrial Internet of Things (IIOT) for supporting new services through interworking and integration with other industries; Integrated Access and Backhaul (IAB) for providing nodes for network service area expansion by supporting wireless backhaul links and access links in an integrated manner; Mobility enhancements including conditional handover and dual active protocol stack (DAPS) handover; and two-step random access for simplifying the random access process (two-step random access channel (RACH) for NR). At the same time, the standardization of system architecture / services for technologies such as the 5G baseline architecture (e.g., service-based architecture or service-based interface) for combining network function virtualization (NFV) and software defined network (SDN) technologies); and mobile edge computing (MEC) for receiving services based on UE location.

[0006] With the commercialization of 5G mobile communication systems, an exponentially growing number of connected devices will be connected to the communication network, and it is therefore expected that enhanced functions and performance of 5G mobile communication systems and integrated operations of connected devices will be necessary. To this end, new research on the following technologies is on the agenda: extended reality (XR) for efficient support of augmented reality (AR), virtual reality (VR), mixed reality (MR), etc.; 5G performance improvement and complexity reduction by utilizing artificial intelligence (AI) and machine learning (ML); AI service support; metaverse service support; and drone communication.

[0007] Wireless communication has been one of the most successful innovations in modern history. Recently, the number of subscribers to wireless communication services has exceeded five billion and continues to grow rapidly. Due to the increasing popularity of smartphones and other mobile data devices (such as tablet computers, "notebook" computers, netbooks, e-book readers, and machine-type devices) among consumers and business people, the demand for wireless data services has increased rapidly. In order to meet the high growth of mobile data services and support new applications and deployments, improvements in radio interface efficiency and coverage are crucial. In order to meet the increased demand for wireless data services since the deployment of 4G communication systems, and to enable various vertical applications, 5G communication systems have been developed and are currently being deployed. Summary of the invention

[0008] Solution to the problem

[0009] A method performed by a user equipment (UE) in a wireless communication system, the method comprising: receiving: first information related to monitoring performance of a machine learning (ML) model for predicting CSI, second information related to transmitting a performance monitoring report, third information related to reception of a CSI reference signal (CSI-RS) for determining true CSI on a cell, and a CSI-RS based on the third information; determining: determining the true CSI based on reception of the CSI-RS, and determining a performance monitoring report of the ML model based on the first information and the true CSI; and transmitting a channel with the performance monitoring report based on the second information.

[0010] A user equipment (UE) in a wireless communication system, the UE comprising: a transceiver; a controller connected to the transceiver and configured to: receive first information related to monitoring performance of a machine learning (ML) model for predicting CSI, receive second information related to transmitting a performance monitoring report, receive third information related to reception of a CSI reference signal (CSI-RS) for determining true CSI on a cell, and receive the CSI-RS based on the third information; and a processor operably connected to the transceiver, determine the true CSI based on reception of the CSI-RS, and determine a performance monitoring report of the ML model based on the first information and the true CSI, wherein the transceiver is further configured to transmit a channel with the performance monitoring report based on the second information.

[0011] A base station in a wireless communication system, the base station comprising: a transceiver; and a controller connected to the transceiver and configured to: transmit first information related to monitoring performance of a machine learning (ML) model for predicting CSI, transmit second information related to transmitting a performance monitoring report, transmit third information related to transmission of a CSI reference signal (CSI-RS) for determining true CSI on a cell, transmit the CSI-RS based on the third information, and receive a channel having a performance monitoring report of the ML model based on the first information and the true CSI based on the second information.

[0012] A method performed by a base station in a wireless communication system, the method comprising: transmitting first information related to monitoring performance of a machine learning (ML) model for predicting CSI, transmitting second information related to transmitting a performance monitoring report, transmitting third information related to transmission of a CSI reference signal (CSI-RS) for determining true CSI on a cell, transmitting the CSI-RS based on the third information, and receiving a channel having a performance monitoring report of the ML model based on the first information and the true CSI based on the second information. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] For a more complete understanding of the present disclosure and its advantages, reference is now made to the following description taken in conjunction with the accompanying drawings, in which like reference numerals represent like parts: Figure 1 An example wireless network according to an embodiment of the present disclosure is shown; Figure 2 An example gNodeB (gNB) according to an embodiment of the present disclosure is shown; Figure 3 An example user equipment (UE) according to an embodiment of the present disclosure is shown; Figure 4A and Figure 4B shows examples of wireless transmit and receive paths according to embodiments of the present disclosure; Figure 5 An example of a transmitter structure for beamforming according to an embodiment of the present disclosure is shown; Figure 6 An example of a graph illustrating artificial intelligence (AI) / machine learning (ML) based CSI feedback according to an embodiment of the present disclosure is shown; Figure 7 An example of a timeline for CSI measurement and reporting according to an embodiment of the present disclosure is shown; Figure 8 An example CSI prediction and a process of transmitting the predicted CSI at the UE and receiving the predicted CSI at the network according to an embodiment of the present disclosure are shown; Fig. 9 An example CSI prediction process performed at a UE according to an embodiment of the present disclosure is shown; Fig.10 A flowchart illustrating an example UE process for CSI prediction according to an embodiment of the present disclosure is shown; Fig.11 An example timeline of CSI measurement window and CSI prediction window indication according to an embodiment of the present disclosure is shown; Fig.12 shows an example timeline indicating an instance of CSI prediction according to an embodiment of the present disclosure; Fig.13 A process of example CSI prediction at a network according to an embodiment of the present disclosure is shown; Fig.14 An example CSI prediction process performed at a network according to an embodiment of the present disclosure is shown; Fig.15 A flowchart illustrating an example UE process for providing assistance information to the network for CSI prediction according to an embodiment of the present disclosure is shown; Fig.16A flowchart illustrating an example UE process for supporting lifecycle management for CSI prediction according to an embodiment of the present disclosure is shown; Fig.17 A flowchart illustrating an example UE process for monitoring CSI prediction performance according to an embodiment of the present disclosure is shown; Fig.18 shows an example timeline of CSI prediction performance monitoring according to an embodiment of the present disclosure; and Fig.19 An example timeline of measuring and filtering CSI prediction performance according to an embodiment of the present disclosure is shown.

[0014] Fig. 20 The structure of a UE according to an embodiment of the present disclosure is shown.

[0015] Fig.21 The structure of a base station according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0016] The present disclosure relates to methods and apparatus for predicting CSI in a cellular system.

[0017] In one embodiment, a method for reporting CSI by a user equipment (UE) is provided. The method includes receiving: first information related to monitoring the performance of a machine learning (ML) model for predicting CSI, second information related to transmitting a performance monitoring report, third information related to the reception of a CSI reference signal (CSI-RS) for determining the true CSI on a cell, and a CSI-RS based on the third information. The method also includes determining the true CSI based on the reception of the CSI-RS, and determining a performance monitoring report of the ML model based on the first information and the true CSI. The method also includes transmitting a channel with a performance monitoring report based on the second information.

[0018] In another embodiment, a UE is provided. The UE includes a transceiver configured to: receive first information related to monitoring the performance of an ML model for predicting CSI, receive second information related to transmitting a performance monitoring report, receive third information related to the reception of a CSI-RS for determining the true CSI on a cell, and receive the CSI-RS based on the third information. The UE also includes a processor operably connected to the transceiver. The processor is configured to: determine the true CSI based on the reception of the CSI-RS, and determine a performance monitoring report of the ML model based on the first information and the true CSI. The transceiver is also configured to transmit a channel with a performance monitoring report based on the second information.

[0019] In yet another embodiment, a base station is provided. The base station includes a transceiver configured to: transmit first information related to monitoring performance of an ML model for predicting CSI, transmit second information related to transmitting a performance monitoring report, transmit third information related to transmission of a CSI-RS for determining true CSI on a cell, transmit the CSI-RS based on the third information, and receive a channel having a performance monitoring report of the ML model based on the first information and the true CSI based on the second information.

[0020] Other technical features may be apparent to those skilled in the art from the following drawings, descriptions and claims.

[0021] Before the following detailed description, it may be advantageous to set forth the definitions of certain words and phrases used throughout this patent document.The term "connection" and its derivatives refer to any direct or indirect communication between two or more elements, whether or not these elements are in physical contact with each other.The terms "transmit", "receive" and "communication" and their derivatives cover direct and indirect communication.The terms "include" and "comprise" and their derivatives mean including but not limited to.The term "or" is inclusive, meaning and / or.The phrase "associated with..." and its derivatives mean including, included in, interconnected with, included in, included in, connected to, connected to, connected to, connected to, connected to, can communicate with, collaborate with, interlace, juxtapose, adjacent to, coupled to, coupled to, have, have characteristics, have a relationship with, or have a relationship with.The term "controller" means any device, system, or part thereof that controls at least one operation.Such a controller can be implemented in hardware or a combination of hardware and software and / or firmware.The function associated with any particular controller can be centralized or distributed, whether local or remote. The phrase "at least one of" when used with a list of items means that different combinations of one or more of the listed items may be used, and only one item in the list may be required. For example, "at least one of A, B, and C" includes any of the following combinations: A; B; C; A and B; A and C; B and C; and A and B and C.

[0022] In addition, the various functions described below may be implemented or supported by one or more computer programs, each of which is formed by a computer-readable program code and embodied in a computer-readable medium. The terms "application" and "program" refer to one or more computer programs, software components, instruction sets, processes, functions, objects, classes, instances, related data or a portion thereof suitable for implementation with suitable computer-readable program code. The phrase "computer-readable program code" includes any type of computer code, including source code, object code and executable code. The phrase "computer-readable medium" includes any type of medium that can be accessed by a computer, such as a read-only memory (ROM), a random access memory (RAM), a hard drive, a compact disk (CD), a digital video disk (DVD) or any other type of memory. "Non-transitory" computer-readable media excludes wired, wireless, optical or other communication links that transmit transient electrical or other signals. Non-transitory computer-readable media include media that can permanently store data, as well as media that can store data and then rewrite data, such as rewritable optical disks or erasable memory devices.

[0023] Definitions for certain other words and phrases are provided throughout this patent document. Those of ordinary skill in the art should understand that in many, if not most instances, such definitions apply to prior, as well as future uses of such defined words and phrases.

[0024] The following discussion Figures 1 to 19 The various non-limiting embodiments used to describe the principles of the present disclosure in this patent document are merely for illustration and should not be interpreted in any way as limiting the scope of the present disclosure. Those skilled in the art will appreciate that the principles of the present disclosure can be implemented in any appropriately arranged system or device.

[0025] In order to meet the increased demand for wireless data services since the deployment of 4G communication systems and to realize various vertical applications, 5G / NR communication systems have been developed and are currently being deployed. 5G / NR communication systems are implemented in higher frequency (millimeter wave) bands (e.g., 28 GHz or 60 GHz bands) to implement higher data rates, or in lower frequency bands (such as 6 GHz) to implement robust coverage and mobility support. In order to reduce the propagation loss of radio waves and increase the transmission distance, beamforming, massive multiple-input multiple-output (MIMO), omnidirectional MIMO (FD-MIMO), array antennas, analog beamforming, and large antenna technologies are discussed in 5G / NR communication systems.

[0026] In addition, in 5G / NR communication systems, development of system network improvements based on advanced small cells, cloud radio access networks (RAN), ultra-dense networks, device-to-device (D2D) communication, wireless backhaul, mobile networks, cooperative communications, coordinated multi-point (CoMP), receiving-side interference cancellation, etc. is underway.

[0027] The discussion of 5G systems and frequency bands associated therewith is for reference, as certain embodiments of the present disclosure may be implemented in 5G systems. However, the present disclosure is not limited to 5G systems or frequency bands associated therewith, and embodiments of the present disclosure may be used in conjunction with any frequency band. For example, aspects of the present disclosure may also be applied to 5G communication systems, 6G, or even higher versions of deployments that may use terahertz (THz) frequency bands.

[0028] The following documents and standard descriptions are hereby incorporated by reference into the present disclosure as if fully set forth herein: [1] 3GPP TS 38.211 v17.2.0, "NR; Physical channels and modulation"; [2] 3GPP TS 38.212 v17.2.0, "NR; Multiplexing and channel coding"; [3] 3GPP TS 38.213 v17.2.0, "NR; Physical layer procedures for control"; [4] 3GPP TS 38.214 v17.2.0, "NR; Physical layer procedures for data"; [5] 3GPP TS 38.215 v17.1.0, "NR; Physical layer measurements"; [6] 3GPP TS 38.331 v17.1.0, "NR; Radio Resource Control (RRC) protocol specification"; [7] 3GPP TS 38.321 v17.1.0, “NR; Medium Access Control (MAC) Protocol Specification”; [8] 3GPP TS 38.133 v17.6.0, “NR; Requirements for support of Radio Resource Management”.

[0029] the following Figures 1 to 3 Various embodiments are described that may be implemented using Orthogonal Frequency Division Multiplexing (OFDM) or Orthogonal Frequency Division Multiple Access (OFDMA) communication techniques in a wireless communication system. Figures 1 to 3 The description is not meant to imply physical or architectural limitations to how different embodiments may be implemented. Different embodiments of the disclosure may be implemented in any suitably arranged communications system.

[0030] Figure 1 An example wireless network 100 is shown in accordance with an embodiment of the present disclosure. Figure 1 The embodiment of the wireless network 100 shown in FIG. 1 is for illustration only. Other embodiments of the wireless network 100 may be used without departing from the scope of the present disclosure.

[0031] like Figure 1 As shown, wireless network 100 includes gNB 101 (e.g., base station, BS), gNB 102, and gNB 103. gNB 101 communicates with gNB 102 and gNB 103. gNB 101 also communicates with at least one network 130 such as the Internet, a proprietary Internet Protocol (IP) network, or other data network.

[0032] The gNB 102 provides wireless broadband access to a network 130 for a plurality of first user equipments (UEs) located within a coverage area 120 of the gNB 102. The plurality of first UEs include: a UE 111, which may be located in a small business; a UE 112, which may be located in an enterprise; a UE 113, which may be a WiFi hotspot; a UE 114, which may be located in a first residence; a UE 115, which may be located in a second residence; and a UE 116, which may be a mobile device such as a cellular phone, a wireless laptop, a wireless PDA, etc. The gNB 103 provides wireless broadband access to the network 130 for a plurality of second UEs within a coverage area 125 of the gNB 103. The plurality of second UEs include UE 115 and UE 116. In some embodiments, one or more of the gNBs 101 to 103 may communicate with each other and with the UEs 111 to 116 using 5G / NR, Long Term Evolution (LTE), Long Term Evolution Advanced (LTE-A), WiMAX, WiFi, or other wireless communication technologies.

[0033] Depending on the network type, the term "base station" or "BS" may refer to any component (or set of components) configured to provide wireless access to a network, such as a transmission point (TP), a transmission-reception point (TRP), an enhanced base station (eNodeB or eNB), a 5G / NR base station (gNB), a macro cell, a femto cell, a WiFi access point (AP), or other devices with wireless functions. A base station may provide wireless access according to one or more wireless communication protocols, for example, 5G / NR 3rd Generation Partnership Project (3GPP) NR, Long Term Evolution (LTE), LTE Advanced (LTE-A), High Speed ​​Packet Access (HSPA), Wi-Fi 802.11a / b / g / n / ac, etc. For convenience, the terms "BS" and "TRP" are used interchangeably in this patent document to refer to network infrastructure components that provide wireless access to remote terminals. In addition, depending on the network type, the term "user equipment" or "UE" may refer to any component, such as a "mobile station", "subscriber station", "remote terminal", "wireless terminal", "reception point", or "user device". For convenience, the terms "user equipment" and "UE" are used in this patent document to refer to a remote wireless device that wirelessly accesses a BS, whether the UE is a mobile device (such as a mobile phone or smartphone) or is generally considered a fixed device (such as a desktop computer or vending machine).

[0034] The dashed lines illustrate the approximate extents of coverage areas 120 and 125, which are shown as generally circular for purposes of illustration and explanation only. It should be clearly understood that coverage areas associated with gNBs, such as coverage areas 120 and 125, may have other shapes, including irregular shapes, depending on the configuration of the gNB and variations in the radio environment associated with natural and man-made obstacles.

[0035] As described in more detail below, one or more of the UEs 111 to 116 include circuits, procedures, or a combination thereof for predicting future CSI in a cellular system. In some embodiments, one or more of the BSs 101 to 103 include circuits, procedures, or a combination thereof for supporting predicting future CSI in a cellular system.

[0036] Although Figure 1 An example of a wireless network is shown, but Figure 1Various changes may be made. For example, wireless network 100 may include any number of gNBs and any number of UEs in any suitable arrangement. In addition, gNB 101 may communicate directly with any number of UEs and provide those UEs with wireless broadband access to network 130. Similarly, each gNB 102-103 may communicate directly with network 130 and provide the UEs with direct wireless broadband access to network 130. In addition, gNBs 101, 102, and / or 103 may provide access to other or additional external networks, such as an external telephone network or other type of data network.

[0037] Figure 2 An example gNB 102 is shown according to an embodiment of the present disclosure. Figure 2 The embodiment of the gNB 102 shown is for illustration purposes only, and Figure 1 gNBs 101 and 103 may have the same or similar configurations. However, gNBs have a variety of configurations, and Figure 2 The scope of this disclosure is not limited to any particular implementation of the gNB.

[0038] like Figure 2 As shown, the gNB 102 includes multiple antennas 205a to 205n, multiple transceivers 210a to 210n, a controller / processor 225, a memory 230, and a backhaul or network interface 235.

[0039] The RF transceivers 210a to 210n receive incoming radio frequency (RF) signals from the antennas 205a to 205n, such as signals transmitted by UEs in the network 100. The transceivers 210a to 210n down-convert the incoming RF signals to generate IF or baseband signals. The IF or baseband signals are processed by receive (RX) processing circuits in the transceivers 210a to 210n and / or the controller / processor 225 to generate processed baseband signals by filtering, decoding and / or digitizing the baseband or IF signals. The controller / processor 225 may further process the baseband signals.

[0040] Transmit (TX) processing circuitry in transceivers 210a to 210n and / or controller / processor 225 receives analog or digital data (such as voice data, network data, email, or interactive video game data) from controller / processor 225. TX processing circuitry encodes, multiplexes, and / or digitizes outgoing baseband data to generate processed baseband or IF signals. Transceivers 210a to 210n up-convert the baseband or IF signals into RF signals that are transmitted via antennas 205a to 205n.

[0041] The controller / processor 225 may include one or more processors or other processing devices that control the overall operation of the gNB 102. For example, the controller / processor 225 may control the reception of uplink (UL) channel signals and the transmission of downlink (DL) channel signals by the transceivers 210a to 210n in accordance with well-known principles. The controller / processor 225 may also support additional functionality, such as more advanced wireless communication functionality. For example, the controller / processor 225 may support beamforming or directional routing operations, in which outgoing / incoming signals from / to multiple antennas 205a to 205n are weighted differently to effectively direct the outgoing signals in a desired direction. As another example, the controller / processor 225 may support methods for supporting prediction of future CSI in a cellular system. The controller / processor 225 may support any of a variety of other functions in the gNB 102.

[0042] The controller / processor 225 is also capable of executing programs and other processes resident in the memory 230, such as to support prediction of future CSI in a cellular system. The controller / processor 225 may move data into or out of the memory 230 as required by the executing process.

[0043] The controller / processor 225 is also connected to a backhaul or network interface 235. The backhaul or network interface 235 allows the gNB 102 to communicate with other devices or systems over a backhaul connection or over a network. The interface 235 may support communication over any suitable wired or wireless connection. For example, when the gNB 102 is implemented as part of a cellular communication system (such as a cellular communication system supporting 5G / NR, LTE, or LTE-A), the interface 235 may allow the gNB 102 to communicate with other gNBs over a wired or wireless backhaul connection. When the gNB 102 is implemented as an access point, the interface 235 may allow the gNB 102 to communicate with a larger network (such as the Internet) over a wired or wireless local area network or over a wired or wireless connection. The interface 235 includes any suitable structure that supports communication over a wired or wireless connection (such as Ethernet or a transceiver).

[0044] Memory 230 is connected to controller / processor 225. A portion of memory 230 may include RAM, and another portion of memory 230 may include flash memory or other ROM.

[0045] Although Figure 2 An example of a gNB 102 is shown, but the Figure 2 For example, gNB 102 may include any number of Figure 2 Each component shown. In addition, Figure 2The various components in may be combined, further subdivided, or omitted, and additional components may be added according to specific needs.

[0046] In this disclosure, italic names of parameters mean that the parameter is provided by higher layers.DL transmissions or UL transmissions may be based on OFDM waveforms, including using a variant of Discrete Fourier Transform (DFT) precoding known as DFT-Spread-OFDM commonly applied to UL transmissions.

[0047] In the present disclosure, a subframe (SF) refers to a transmission time unit of the LTE RAT, and a slot refers to a transmission time unit of the NR RAT. For example, the slot duration may be a fraction of the SF duration. NR may use a DL or UL slot structure that is different from the LTE SF structure. The differences may include a structure for transmitting a physical downlink control channel (PDCCH), a location and structure of a demodulation reference signal (DM-RS), a transmission duration, etc. In addition, an eNB refers to a base station that provides services for UEs operating using the LTE RAT, and a gNB refers to a base station that provides services for UEs operating using the NR RAT. An exemplary embodiment examines the same set of parameters for transmission using the LTE RAT and the NR RAT, which includes a subcarrier spacing (SCS) configuration and a cyclic prefix (CP) length of an OFDM symbol. In this case, the OFDM symbol of the LTE RAT is the same as that of the NR RAT, the subframe is the same as the slot, and for brevity, the term slot is used in the remainder of the present disclosure.

[0048] A unit for DL ​​signaling or for UL signaling on a cell is called a slot and may include one or more symbols. A bandwidth (BW) unit is called a resource block (RB). One RB includes multiple subcarriers (SCs). For example, a slot may have a duration of one millisecond, and an RB may have a bandwidth of 180 kHz and include 12 SCs with an inter-SC spacing of 15KHz. Subcarrier spacing (SCS) may be configured by SCS. Determined as kHz. The unit of one subcarrier on one symbol is called a resource element (RE). The unit of one RB on one symbol is called a physical RB (PRB).

[0049] MIMO technology plays an important role in improving system throughput in NR and LTE, and this role will continue and expand further in the next generation of wireless technologies.

[0050] The antenna ports are defined so that the channel that delivers another symbol on the same antenna port can be inferred from the channel that delivered the previous symbol of the antenna port. There is not necessarily a one-to-one correspondence between antenna ports and antenna elements, and multiple antenna elements can be mapped to one antenna port.

[0051] Figure 3 An example UE 116 is shown in accordance with an embodiment of the present disclosure. Figure 3 The embodiment of UE 116 shown is for illustration only, and Figure 1 UEs 111 to 115 may have the same or similar configurations. However, UEs have various configurations, and Figure 3 The scope of the present disclosure is not limited to any particular implementation of the UE.

[0052] like Figure 3 As shown, UE 116 includes antenna 305, transceiver 310, and microphone 320. UE 116 also includes speaker 330, processor 340, input / output (I / O) interface (IF) 345, input 350, display 355, and memory 360. Memory 360 includes operating system (OS) 361 and one or more applications 362.

[0053] The transceiver 310 receives incoming RF signals transmitted by the gNB of the wireless network 100 from the antenna 305. The transceiver 310 downconverts the incoming RF signals to generate an intermediate frequency (IF) or baseband signal. The IF or baseband signal is processed by the RX processing circuitry and / or the processor 340 in the transceiver 310 by filtering, decoding and / or digitizing the baseband or IF signal to generate a processed baseband signal. The RX processing circuitry sends the processed baseband signal to the speaker 330 (such as for voice data) or is processed by the processor 340 (such as for web browsing data).

[0054] The TX processing circuitry in the transceiver 310 and / or processor 340 receives analog or digital voice data from the microphone 320 or other outgoing baseband data (such as web data, email, or interactive video game data) from the processor 340. The TX processing circuitry encodes, multiplexes, and / or digitizes the outgoing baseband data to generate a processed baseband or IF signal. The transceiver 310 up-converts the baseband or IF signal into an RF signal that is transmitted via the antenna 305.

[0055] The processor 340 may include one or more processors or other processing devices and execute the OS 361 stored in the memory 360 to control the overall operation of the UE 116. For example, the processor 340 may control the transceiver 310 to receive DL channel signals and transmit UL channel signals according to well-known principles. In some embodiments, the processor 340 includes at least one microprocessor or microcontroller.

[0056] The processor 340 is also capable of executing other processes and programs resident in the memory 360. For example, the processor 340 can execute the process for predicting future CSI in a cellular system as described in an embodiment of the present disclosure. The processor 340 can move data into or out of the memory 360 as needed to execute the process. In some embodiments, the processor 340 is configured to execute the application 362 based on the OS 361 or in response to a signal received from the gNB or operator. The processor 340 is also connected to the I / O interface 345, which provides the UE 116 with the ability to connect to other devices such as laptops and handheld computers. The I / O interface 345 is the communication path between these accessories and the processor 340.

[0057] Processor 340 is also connected to input 350 (which includes, for example, a touch screen, a keypad, etc.) and display 355. An operator of UE 116 may use input 350 to enter data into UE 116. Display 355 may be a liquid crystal display, a light emitting diode display, or other display capable of presenting text and / or at least limited graphics (such as from a website).

[0058] The memory 360 is connected to the processor 340. A portion of the memory 360 may include a random access memory (RAM), and another portion of the memory 360 may include a flash memory or other read-only memory (ROM).

[0059] Although Figure 3 An example of UE 116 is shown, but the Figure 3 Make various changes. For example, Figure 3 Various components in the embodiment may be combined, further subdivided, or omitted, and additional components may be added as required. As a specific example, processor 340 may be divided into multiple processors, such as one or more central processing units (CPUs) and one or more graphics processing units (GPUs). In another example, transceiver 310 may include any number of transceivers and signal processing chains and may be connected to any number of antennas. In addition, although Figure 3 The UE 116 is shown configured as a mobile phone or smart phone, but the UE may be configured to operate as other types of mobile or stationary devices.

[0060] Figure 4A and Figure 4B Examples of wireless transmit and receive paths 400 and 450, respectively, are shown in accordance with embodiments of the present disclosure. For example, transmit path 400 may be described as being implemented in a gNB (such as gNB 102), while receive path 450 may be described as being implemented in a UE (such as UE 116). However, it will be understood that receive path 450 may be implemented in a gNB and transmit path 400 may be implemented in a UE. In some embodiments, receive path 450 is configured to support prediction of future CSI in a cellular system, as described in embodiments of the present disclosure.

[0061] like Figure 4A As shown, the transmit path 400 includes a channel coding and modulation block 405, a serial to parallel (S to P) block 410, an inverse fast Fourier transform (IFFT) block of size N 415, a parallel to serial (P to S) block 420, an add cyclic prefix block 425, and an up converter (UC) 430. The receive path 450 includes a down converter (DC) 455, a remove cyclic prefix block 460, an S to P block 465, a fast Fourier transform (FFT) block of size N 470, a parallel to serial (P to S) block 475, and a channel decoding and demodulation block 480.

[0062] In the transmit path 400, the channel coding and modulation block 405 receives a set of information bits, applies coding (such as low density parity check (LDPC) coding), and modulates the input bits (such as with quadrature phase shift keying (QPSK) or quadrature amplitude modulation (QAM)) to generate a sequence of frequency domain modulation symbols. The serial to parallel block 410 converts (such as demultiplexes) the serially modulated symbols into parallel data to generate N parallel symbol streams, where N is the IFFT / FFT size used in the gNB 102 and UE 116. The size-N IFFT block 415 performs an IFFT operation on the N parallel symbol streams to generate a time domain output signal. The parallel to serial block 420 converts (such as multiplexes) the parallel time domain output symbols from the size-N IFFT block 415 to generate a serial time domain signal. The add cyclic prefix block 425 inserts a cyclic prefix into the time domain signal. Upconverter 430 modulates (such as upconverts) the output of add cyclic prefix block 425 to RF frequency for transmission via a wireless channel. The signal may also be filtered at baseband prior to conversion to RF frequency.

[0063] like Figure 4BAs shown, the down converter 455 down-converts the received signal to the baseband frequency, and the remove cyclic prefix block 460 removes the cyclic prefix to generate a serial time domain baseband signal. The serial to parallel block 465 converts the time domain baseband signal to a parallel time domain signal. The FFT block 470 of size N performs an FFT algorithm to generate N parallel frequency domain signals. The (P to S) block 475 converts the parallel frequency domain signals into a sequence of modulated data symbols. The channel decoding and demodulation block 480 demodulates and decodes the modulated symbols to recover the original input data stream.

[0064] Each of gNBs 101-103 may implement a transmit path 400 similar to that for transmitting in the downlink to UEs 111-116 and may implement a receive path 450 similar to that for receiving in the uplink from UEs 111-116. Similarly, each of UEs 111-116 may implement a transmit path 400 for transmitting in the uplink to gNBs 101-103 and may implement a receive path 450 for receiving in the downlink from gNBs 101-103.

[0065] Can be implemented using hardware only or a combination of hardware and software / firmware Figure 4A and Figure 4B As a specific example, Figure 4A and Figure 4B At least some of the components in can be implemented in software, while other components can be implemented by configurable hardware or a mixture of software and configurable hardware. For example, FFT block 470 and IFFT block 415 can be implemented as a configurable software algorithm, wherein the value of size N can be modified according to the implementation.

[0066] In addition, although described as using FFT and IFFT, this is merely an illustrative approach and should not be construed as limiting the scope of the present disclosure. Other types of transforms may be used, such as discrete Fourier transform (DFT) and inverse discrete Fourier transform (IDFT) functions. It should be understood that the value of the variable N may be any integer (such as 1, 2, 3, 4, etc.) for the DFT and IDFT functions, and the value of the variable N may be any integer (such as 1, 2, 4, 8, 16, etc.) as a power of two for the FFT and IFFT functions.

[0067] although Figure 4A and Figure 4B Examples of wireless transmission path 400 and wireless reception path 450 are shown separately, but can be Figure 4A and Figure 4B Make various changes. For example, Figure 4A and Figure 4BThe various components in the can be combined, further subdivided or omitted, and additional components can be added according to specific needs. Figure 4A and Figure 4B It is intended to illustrate examples of the types of transmit and receive paths that may be used in a wireless network. Any other suitable architecture may be used to support wireless communications in a wireless network.

[0068] Figure 5 An example of a transmitter structure 500 for beamforming according to an embodiment of the present disclosure is shown. In certain embodiments, one or more of the gNB 102 or the UE 116 includes the transmitter structure 500. For example, one or more of the antenna 205 and its associated system, or the antenna 305 and its associated system may be included in the transmitter structure 500. This example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0069] Thus, embodiments of the present disclosure recognize that Rel-14 LTE and Rel-15 NR support up to 32 channel state information reference signal (CSI-RS) antenna ports, which enables an eNB or gNB to be equipped with a large number of antenna elements (such as 64 or 128). Multiple antenna elements can then be mapped to one CSI-RS port. For mmWave bands, while the number of antenna elements can be larger for a given form factor, the number of CSI-RS ports that can correspond to the number of digital precoding ports may be larger due to factors such as Figure 5 The analog beamforming technology shown in FIG. 5 is limited by hardware constraints such as the feasibility of installing a large number of analog-to-digital converters (ADCs) / digital-to-analog converters (DACs) at mmWave frequencies. One CSI-RS port can then be mapped to a large number of antenna elements that can be controlled by a set of analog phase shifters 501. One CSI-RS port can then correspond to a subarray that produces a narrow analog beam through analog beamforming 505. This analog beam can be configured to sweep a wider angular range 520 by changing the phase shifter set across symbols or time slots / subframes. The number of subarrays (equal to the number of RF chains) is proportional to the number of CSI-RS ports. N CSI-PORT The digital beam forming unit 510 performs cross N CSI-PORT The digital precoding can be varied across frequency subbands or resource blocks, although the analog beams are wideband (and therefore not frequency selective). Receiver operation can be similarly conceived.

[0070] because Figure 5The transmitter structure 500 utilizes multiple analog beams for transmission and reception (where one or a small number of analog beams are selected from a large number of analog beams, for example, after a training duration that is performed occasionally or periodically), so the term "multi-beam operation" is used to refer to the entire system aspect. For illustrative purposes, this includes indicating an assigned DL or UL TX beam (also referred to as "beam indication"), measuring at least one reference signal for calculating and performing beam reporting (also referred to as "beam measurement" and "beam reporting", respectively), and receiving DL or UL transmissions via selecting a corresponding RX beam. Figure 5 The system is also applicable to higher frequency bands, such as >52.6 GHz (also known as frequency range 4 or FR4). In this case, the system can only use analog beams. Due to the O2 absorption loss near 60 GHz (about 10 dB additional loss per 100 m distance), a large number of narrow analog beams (and therefore a larger number of radiators in the array) are required to compensate for the additional path loss.

[0071] The text and drawings are provided as examples only to help the reader understand the present disclosure. They are not intended and should not be interpreted as limiting the scope of the present disclosure in any way. Although certain embodiments and examples have been provided, it will be apparent to those skilled in the art based on the disclosure herein that changes may be made to the embodiments and examples shown without departing from the scope of the present disclosure. The transmitter structure 500 for beamforming is for illustration only and other embodiments may be used without departing from the scope of the present disclosure.

[0072] In order to achieve digital precoding, efficient design of CSI-RS is a key factor. To this end, three types of CSI reporting mechanisms corresponding to three types of CSI-RS measurement behaviors are supported in Rel.13 LTE: 1) "Type A" CSI report, which corresponds to non-precoded CSI-RS; 2) "Type B" report, with K=1 CSI-RS resources corresponding to UE-specific beamformed CSI-RS; 3) "Type B" report, with K>1 CSI-RS resources corresponding to cell-specific beamformed CSI-RS. For non-precoded (NP) CSI-RS, a cell-specific one-to-one mapping between CSI-RS ports and transceiver units (TXRUs) is utilized. Here, different CSI-RS ports have the same wide beamwidth and direction, and therefore have a coverage range that is roughly the width of the cell. For beamformed CSI-RS, beamforming operations (whether cell-specific or UE-specific) are applied to non-zero power (NZP) CSI-RS resources (e.g., including multiple ports). Here, (at least at a given time / frequency) the CSI-RS ports have narrower beamwidths and thus do not have cell-wide coverage, and (at least from the eNB's perspective) at least some CSI-RS port-resource combinations have different beam directions. In NR, the basic principles remain the same.

[0073] In scenarios where the DL long-term channel statistics can be measured via the UL signal at the serving gNB, UE-specific beamformed CSI-RS can be easily used. This is usually feasible when the UL-DL duplex distance is small enough. However, when this condition does not hold, some UE feedback is beneficial for the gNB to obtain an estimate of the DL long-term channel statistics (or any representation thereof). To facilitate such a process, a first beamformed CSI-RS is transmitted with a periodicity of T1 (ms) and a second NP CSI-RS is transmitted with a periodicity of T2 (ms), where T1 ≤ T2. This approach is called hybrid CSI-RS. The implementation of hybrid CSI-RS depends heavily on the definition of the CSI process and the NZP CSI-RS resources.

[0074] One of the important components of MIMO transmission schemes is accurate CSI acquisition at the gNB (or TRP). Especially for multi-user (MU)-MIMO, the availability of accurate CSI is necessary in order to guarantee high MU performance. For time division duplex (TDD) systems, CSI can be acquired using SRS transmission relying on channel reciprocity. On the other hand, for frequency division duplex (FDD) systems, CSI can be acquired using CSI-RS transmission from the gNB and CSI acquisition and feedback from the UE. In LTE prior to Rel. 13, for FDD systems, assuming single-user (SU) transmission from the eNB, the CSI feedback framework was "implicit" in the form of channel quality information (CQI) / precoding matrix indicator (PMI) / rank indicator (RI) (and CSI-RS indicator (CRI) in Rel. 13) derived from the codebook. Due to the inherent SU assumption in deriving CSI, this implicit CSI feedback is not sufficient for MU transmissions. On the other hand, the NR system has been designed to be more MU-centric since its first version, with a high-resolution Type-II codebook in addition to the low-resolution Type-I codebook.

[0075] In current networks, the application of AI / ML-based methods is mainly limited to the network layer. The O-RAN Alliance has defined a virtualized RAN with open interfaces and network intelligence with entities such as non-real-time (RT) RAN Intelligent Controller (RIC) and near-RT RIC. Non-RT RIC is a logical function capable of non-real-time control and optimization of RAN elements and resources, which manages the overall AI / ML workflow of the O-RAN network, including model training, reasoning, and updates. Near-RT RIC is a logical function capable of near-real-time control and optimization of RAN elements and resources via fine-grained data collection and operation on the RAN interface. On the other hand, 3GPP defined the Network Data Analysis Function (NWDAF) for network slice management in Rel-15 and further enhanced it in Rel-16 and Rel-17. 3GPP also defines a functional framework for RAN intelligence implemented through data collection.

[0076] It is expected that AI / ML methods will be applied to various cellular system air interface designs, including CSI compression / recovery, future CSI prediction, learning-based channel estimation, channel coding, and modulation, to name a few. Common physical layer algorithms have been derived based on simplifying assumptions such as linear system models, additive white Gaussian noise (AWGN) channels, etc. By leveraging AI / ML methods, optimal algorithms can be developed for more practical system assumptions such as nonlinear and fading channels.

[0077] It is also expected that, depending on the use case, not only system performance such as throughput, spectral efficiency, and latency can be improved, but also complexity, reliability, and overhead, etc. In addition, it is possible to optimize not only in a segmented manner for a given transmitter / receiver processing function, but also in an end-to-end manner, including the entire transmitter / receiver processing chain. Therefore, it is expected that the scope of application of AI / ML in cellular systems will continue to expand.

[0078] Figure 6 An example of a graph 600 showing AI / ML based CSI feedback according to an embodiment of the present disclosure is shown. For example, the AI / ML based CSI feedback shown in graph 600 may be Figure 1 The example is performed between the gNB 102 and / or the network 130 and the UE 116 in the wireless network 100. This example is for illustration only and may be used without departing from the scope of the present disclosure.

[0079] The UE derives CSI from DL CSI-RS measurements, compresses the CSI using an AI / ML based CSI encoder, and sends the output of the CSI encoder as a CSI report to the network 130. The network 130 receives the CSI report from the UE 116, uses it as input to the AI / ML based CSI decoder, and reconstructs the CSI from the UE 116. AI / ML based CSI feedback can be considered as a candidate for the next generation CSI acquisition method, which can replace the current method based on DFT basis selection and precoding matrix expression via a linear combination of a selected basis set (such as Type-II codebook in NR).

[0080] Even if the adopted CSI feedback scheme can perfectly transmit the CSI derived at the UE 116 for the current CSI reference resource to the network 130 without any information loss, the validity of the CSI received at the network 130 is still affected by channel aging.

[0081] Figure 7 2 shows an example of a timeline 700 for CSI measurement and reporting according to an embodiment of the present disclosure. For example, the timeline 700 for CSI measurement and reporting may be Figure 3 UE 116 utilizes. This example is for illustration only and may be used without departing from the scope of the present disclosure.

[0082] Use time slot n ref The CSI reference resource at n is used to measure the channel, and considering the processing time required for UE 116 to perform channel measurement and prepare CSI reports, based on ref The CSI derived from the channel measured at the CSI reporting time slot n after a few time slots repis sent to the network 130. Δ represents n ref With n rep The channel aging effect means that even in time slot n rep At the time the CSI is received at the network 130, the CSI is also out of date by a duration Δ. This captures the time-varying channel characteristics and processing delays inherent in real systems. To mitigate this degradation caused by channel aging, future channel prediction or channel tracking is important, where the overall goal is to proactively predict what the actual channel state at the time of use may be based on the available observed channels. For example, Figure 7 In the example, we can use n ref channel measurements at n to predict future instances (e.g., n rep ) of the CSI, and at n rep The predicted CSI is reported at 100 Mbps, thereby mitigating channel aging and providing new information upon reception. This prediction is possible because the wireless channel is essentially a function of its surroundings; therefore, the channel samples are correlated in time.

[0083] CSI prediction may be performed at the UE or the network, which may depend on the availability of information for prediction at the UE 116 or at the network 130, for example due to TDD / FDD system operation. Therefore, it is necessary to design methods and apparatus for CSI prediction, assuming that when the prediction is performed at the UE and when the prediction is performed at the network.

[0084] Embodiments of the present disclosure recognize that when CSI prediction is performed at a UE, it is necessary to indicate to the UE 116 one or more instances for which CSI is derived and reported, including current instances and future instances.

[0085] Embodiments of the present disclosure recognize that when CSI prediction is performed at the UE for more than one instance and reported to the network 130, there is a need to enhance CSI report message construction to reduce feedback overhead.

[0086] Embodiments of the present disclosure recognize that when CSI prediction is performed at the network, a list of beneficial assistance information feedback from the UE 116 to the network 130 needs to be indicated to the UE 116 to assist in CSI prediction at the network 130 .

[0087] The present disclosure relates to communication systems. The present disclosure relates to defining functions and processes to support prediction of future CSI in a cellular system.

[0088] The present disclosure also relates to indicating to a UE one or more instances for which a CSI report is to be constructed, including a current instance and future instances.

[0089] The present disclosure also relates to enhancing CSI reporting message construction when reporting CSI for more than one future instance.

[0090] The present disclosure also relates to indicating assistance information for CSI prediction to a UE and receiving assistance information for CSI prediction from a UE.

[0091] The text and drawings are provided as examples only to help the reader understand the present disclosure. They are not intended and should not be interpreted as limiting the scope of the present disclosure in any way. Although certain embodiments and examples have been provided, it will be understood by those skilled in the art based on the disclosure herein that the embodiments and examples shown may be changed without departing from the scope of the present disclosure.

[0092] The flowcharts herein illustrate exemplary methods that can be implemented according to the principles of the present disclosure, and various changes can be made to the methods shown in the flowcharts herein. For example, although shown as a series of steps, the individual steps in each figure can overlap, occur in parallel, occur in different orders, or occur multiple times. In another example, a step can be omitted or replaced by another step.

[0093] Embodiments of the present disclosure related to predicting future CSI performed at a UE or at a network are summarized below and are set forth further fully herein.

[0094] • Methods and apparatus indicating one or more instances of reporting CSI, including current instances and future instances.

[0095] • Methods and apparatus for reducing CSI feedback overhead when reporting CSI for more than one instance.

[0096] • Methods and apparatus for receiving auxiliary information (eg, channel environment, multipath delay / Doppler profile) from a UE for CSI prediction at the network 130.

[0097] Figure 8 800 for example CSI prediction and transmission of predicted CSI at a UE and reception of predicted CSI at a network according to an embodiment of the present disclosure. For example, the process 800 for example CSI prediction at a UE may be performed at Figure 1 The example is performed between the gNB 102 and / or the network 130 and the UE 116 in the wireless network 100. This example is for illustration only and may be used without departing from the scope of the present disclosure.

[0098] The process starts at 810, UE uses the time slot n ref The channel measured at the CSI reference resource at , and possibly by {n ref -W m ,…,n ref}, with W mA set of past channel measurements for a measurement window duration denoted by . In 820, a CSI predictor is then used for the past channel measurements at UE 116 to predict CSI for future time instances. The predicted future CSI instance is given by {n ref +o,…,n ref +o+W p}, where o is the time slot n ref The time offset of the CSI reference resource in and W p is the time window for prediction. As described above, the CSI prediction can be not only for a single future instance, but also for multiple instances (including current instances and future instances) indicated from the network 130 to the UE 116, which has been disclosed herein. In 830, the predicted future CSI is then encoded using a CSI encoder at the UE 116. The encoded CSI is sent over the air as CSI feedback. In 840, the CSI decoder at the gNB 102 and / or the network 130 then decodes the predicted future CSI based on the received CSI feedback. In 850, the predicted future CSI is then available at the gNB 102 and / or the network 130.

[0099] Fig. 9 FIG. 9 shows an example CSI prediction process 900 performed at a UE according to an embodiment of the present disclosure. For example, the process 900 may be performed by Figure 1 The UE 116 and the gNB 102 and / or the network 130 in the wireless network 100 are performed. This example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0100] The process begins at 910, where the network provides parameters and / or models for CSI prediction to the UE. In 920, the network 130 then provides reference signals for channel measurement and for the UE to perform channel measurement. In 930, the UE 116 then performs CSI prediction for future instances. In 940, the UE 116 then sends a CSI report to the network 130 in accordance with an embodiment of the present disclosure.

[0101] Once the UE 116 predicts the CSI for future instances, the information can be encoded into a CSI report and sent to the network 130 via various methods. In one example, the CSI report message is encoded using a common Type-I or Type-II codebook. In another example, the CSI report message is encoded using a common Type-I or Type-II codebook. Figure 6 In another example, as shown in FIG. Figure 6 The AI / ML based CSI compression method shown is trained to generate CSI encoder output for future predicted CSI, rather than Figure 8 As shown in FIG. 1 , the CSI is first predicted and then compressed in two stages. When CSI is reported for more than one instance (including the current instance and future instances), compression can be performed in the time domain as well as the spatial and frequency domains by using an enhanced Type-II codebook with Doppler domain basis selection or a three-dimensional joint compression method based on AI / ML. Alternatively, CSI reports for multiple instances can be encoded separately and transmitted to the network 130 in one report or multiple separate reports.

[0102] Fig.10 1 shows a flow chart of an example UE process 1000 for CSI prediction according to an embodiment of the present disclosure. For example, the process 1000 for CSI prediction may be Figure 1 Any of the UEs 111 to 116 (such as Figure 3 UE 116) and the corresponding process may be performed by Figure 1 The example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0103] The process begins at 1010, where parameters and / or models for future CSI prediction are provided to the UE from the network 130. Parameters that may be provided by the network 130 to the UE 116 include those related to the CSI measurement window, the CSI prediction window or future instance set for prediction, the CSI report format, and the prediction model, including but not limited to AI / ML-based CSI prediction, prediction based on an extended Kalman filter (EKF), etc. In 1020, the UE 116 then measures reference signals transmitted from the network 130. Reference signals include CSI-RS as well as synchronization signal blocks (SSBs), tracking reference signals (TRSs), and / or any signal transmitted from the network 130 that the UE 116 may use to estimate the channel. In 1030, the UE 116 then calculates the future CSI based on the parameters and / or models provided from the network 130. In 1040, the UE 116 sends a CSI report including the predicted future CSI feedback to the network 130. The CSI feedback report format follows the instructions from the network 130 provided in 1010 .

[0104] Fig.11 An example timeline 1100 for CSI measurement window and CSI prediction window indication according to an embodiment of the present disclosure is shown. For example, the timeline 1100 for CSI measurement window and CSI prediction window indication may be Figure 3 UE 116 utilizes. This example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0105] The UE may predict future CSI based on the measurement window indicated by the network 130. In one example, the measurement window may be indicated to the UE 116 using a duration and an optional offset from the CSI reference resource, e.g., {n ref -W m ,…,n ref}, where W m is the measurement window duration and no offset is configured in this example.

[0106] The prediction window for UE 116 to predict future CSI may be indicated to the UE by network 130. In one example, the prediction window may be indicated to UE 116 using a duration and an offset from a CSI reference resource, e.g., {n ref +o,…,n ref +o+W p}, where W p is the prediction window duration, and o is the prediction start offset from the CSI reference source. Various configurations are possible. For example, if W p If and o are both set to zero, the prediction window indicates the fallback to the ref The traditional feedback scheme for reporting CSI for CSI reference resources is as follows. p If W is set to zero, but o is set to a positive integer, the prediction window is reduced to the indication of a single future instance. p If and o are both set to positive integers, the prediction window indicates the duration W used for CSI prediction. p From n ref Similar to the example in this article, the duration and relative to the CSI reporting instance n can be used rep The offset, that is, {n rep +o,…,n rep +o+W p} to indicate the prediction window to UE 116. In this case, the offset can take any integer value, including negative and zero values.

[0107] Fig.12 An example timeline 1200 indicating an instance of CSI prediction according to an embodiment of the present disclosure is shown. For example, the timeline 1200 for indicating an instance of CSI prediction may be Figure 3 UE 116 utilizes. This example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0108] The prediction start offset o, the prediction interval I, and the number of instances to be used for prediction may be indicated to the UE by the network 130. In the example timeline herein, the start offset from the CSI reference resource is set to 6 slots, the prediction interval is set to 2 slots, and the number of future CSI reports is set to 3. Thus, the UE 116 will generate a prediction report for a set of future instances {n ref +6,n ref +8,n ref +10} to predict CSI. Similarly, we can rep to indicate the starting offset.

[0109] Alternatively, a set of offset values ​​may be indicated by network 130 to UE 116, the set of offset values ​​indicating future instances to be used for CSI prediction. For example, network 130 may indicate a set of offset values ​​to UE 116, such as {o 1 , o 2 , o 3}, and UE 116 will respond to {n ref + o 1 , n ref + o 2 , n ref + o 3} to predict CSI. Similarly, it can be reported with respect to CSI report instance n rep to indicate the starting offset.

[0110] For aperiodic CSI reporting, all or part of the information indicating future instances for prediction may be conveyed in the PDCCH that provides downlink control information (DCI) that triggers the CSI report. For example, the DCI that triggers the aperiodic CSI report may include one or more offset values ​​indicating future instances for CSI prediction. Similarly, the DCI that triggers the aperiodic CSI report may include a prediction start offset o, a prediction interval I, and the number of instances for prediction. Similarly, the DCI that triggers the aperiodic CSI report may include a prediction window duration and a start offset.

[0111] A plurality of configurations regarding future instances for prediction may be provided to the UE via high-layer signaling by the network 130 using any of the methods disclosed herein. The network 130 notifies the UE 116 of the configuration index via a PDCCH providing a DCI format. As an example, a first configuration indicates a set of future instances, for example, by signaling a set of offset values ​​or a tuple {offset, interval, number}, and a second configuration indicates a current instance, i.e., n ref A configuration index to be followed when preparing a CSI report between the first configuration and the second configuration may be indicated from the network 130 to the UE 116 via DCI.

[0112] When a UE reports CSI for more than one instance (including the current instance and / or future instances), the CSI report for each instance may be independent, or the CSI-RS report for one instance may be related to another instance. ref + o 1 , n ref + o 2} instance to report CSI, then ref + o 1 and n ref + o 2 Elements of the CSI report, including but not limited to CQI, PMI, CRI, SS / PBCH resource block indicator (SSBRI), LI, RI, L1 reference signal received power (RSRP), are encoded separately, resulting in separate and independent reports.

[0113] Alternatively, UE 116 may specify a specific instance (eg, n ref + o 2 ) relative to another instance (e.g., n ref + o 1 ) to generate a CSI report. In one example, if UE 116 generates a CSI report for time slot n ref + o 1 The derived CQI index is c 1 And for time slot n ref + o 2 The derived CQI index is c 2 , then UE 116 can use the reduced number of bits to allocate time slot n ref + o 2 The CQI index is reported as c 2 -c 1 In one example, UE 116 may reuse time slot n. ref + o 1 The PMI report selects a spatial domain and / or frequency domain basis set to report in time slot n ref + o 2 PMI, and can be targeted at time slot n ref + o 2 Separately, only the amplitude and / or phase coefficients of the linear combination of the basis set are reported. In another example, UE 116 can reuse the time slot n ref + o 1 Select a set of spatial and / or frequency domain bases to report time slot n ref + o 2 PMI, and using the reduced number of digits relative to n ref + o1 The coefficients used to report time slot n ref + o 2 In one example, UE 116 may report a single RI that is generally applicable to a set of instances reported by the UE. In another example, UE 116 may report a single RI with respect to time slot n. ref + o 1 RI to report time slot n ref + o 2 In one example, for any CSI component measured in dBm, such as CQI, CRI, etc., UE 116 may calculate the difference RI value relative to time slot n. ref + o 1 The amount to report time slot n ref + o 2 The amount is reduced, i.e., differential dBm values ​​are reported using a reduced number of bits.

[0114] If time slot n ref + o 2 Any CSI component with time slot n ref + o 1 If the difference between the reference at and is less than a certain threshold, the UE 116 may skip reporting the component or time slot n. ref + o 2 Such thresholds may be set individually for each of CQI, PMI, CRI, SSBRI, LI, RI, L1-RSRP, and may be provided by the network 130 to the UE 116 or defined by the specification. In one example, if for time slot n ref + o 2 The predicted CQI for time slot n ref + o 1 If the measured / predicted CQI is within a certain range, the UE 116 may skip transmitting in time slot n. ref + o 2 In other words, if |CQI t1 –CQI ref | ≤T, the UE skips the CSI report of time slot t1, where the threshold T can be provided to UE 116 by network 130.

[0115] Fig.13 FIG. 1 shows a process 1300 of example CSI prediction at a network according to an embodiment of the present disclosure. For example, the process 1300 may be performed at Figure 1 The example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0116] In 1310, past channel measurements are made at the UE. In 1320, the channel parameter extractor then extracts the auxiliary information and the CSI encoder performs the ref The auxiliary information and the encoded CSI are both sent over the air to the network 130. In 1330, the encoded CSI is then decoded by the CSI decoder. In 1340, the auxiliary information and the CSI predictor are then received. ref At 1350, the predicted future CSI is then at network 130.

[0117] The UE provides various auxiliary information for CSI prediction to the network 130, and the network 130 performs CSI prediction using the auxiliary information received from the UE 116. The UE uses the auxiliary information received in the time slot n ref The channel measured at the CSI reference resource at , and possibly using a measurement window duration W m By {n ref -W m ,…,n ref} to derive auxiliary information feedback to the network 130. The derived auxiliary information can be compared with the CSI reference resource n ref The CSI feedbacks of the UE and the UE may be transmitted to the network 130 separately, or they may be transmitted to the network 130 jointly.

[0118] Fig.14 1400 is shown as an example CSI prediction process performed at a network according to an embodiment of the present disclosure. For example, the example CSI prediction process 1400 at a network may be performed by Figure 1 The UE 116 and the gNB 102 and / or the network 130 in the wireless network 100 are performed. This example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0119] The process begins at 1410, where the network provides the UE with a list of assistance information requested for CSI prediction at the gNB and / or the network. In 1420, the network 130 then provides a reference signal to the UE 116 for channel measurement and performs CSI prediction based on the assistance information received from the UE 116. In 1430, the UE 116 then performs channel measurement and calculates the assistance information. In 1440, the UE 116 then sends the assistance information feedback to the network 130.

[0120] Fig.15 1 shows a flow chart of an example UE process 1500 for providing auxiliary information to a network for CSI prediction according to an embodiment of the present disclosure. For example, the process 1500 for providing auxiliary information to a network for CSI prediction may be performed by Figure 1 Any one of the UEs 111 to 116 of the present invention may perform the above process, and the corresponding process may be performed by Figure 1 The example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0121] The process starts at 1510, where the network 130 provides a list of auxiliary information to the UE to send to the network 130 for CSI prediction. The auxiliary information may include the channel environment perceived by the UE, multipath delay / Doppler profile, etc. The network 130 may also provide a list of auxiliary information to the UE for CSI prediction via a time duration and an optional offset to the CSI reference resource (e.g., {n ref -W m ,…,n ref}) or an indication of an optional offset from the CSI reporting instance is provided to the UE 116, similar to Fig.11 As shown. In 1520, UE 116 then measures reference signals transmitted from network 130, such as CSI-RS. Reference signals include CSI-RS as well as SSB, TRS, and / or any signal transmitted from network 130 that UE 116 can use to estimate the channel. In 1530, UE 116 then calculates auxiliary information based on the list provided from network 130. In 1540, UE 116 then sends auxiliary information for CSI prediction to network 130.

[0122] The UE provides various auxiliary information for CSI prediction to the network 130, including but not limited to the channel environment, Doppler profile and / or multipath profile perceived by the UE. In one example, the channel environment perceived by the UE 116 may include indications of the following items: UMa / UMi / InH / suburban, clutter / presence of congestion / density / severity, LOS / NLOS indication, indoor / outdoor indication, in-vehicle indication, in-building indication, mobility represented by speed or speed category, for example, pedestrian / vehicle / high-speed train, etc. In one example, the Doppler profile may include Doppler spread, Doppler shift, relative Doppler shift. In one example, the multipath profile may include delay spread, and each path weight, delay, and Doppler value of each signal propagation path. The network 130 may provide a threshold value of signal strength to the UE 116, so that for paths with strength greater than the threshold, the weight, delay, and Doppler value are reported to the network 130. The strength may be expressed in terms of the amplitude or power of the signal. The strength may be measured by averaging the values ​​over the subcarriers and / or symbols carrying the reference signal, or may be considered as the maximum value over the subcarriers and / or symbols carrying the reference signal.

[0123] Embodiments of the present disclosure recognize that the selection of an appropriate CSI prediction model may depend on the channel environment and processing capabilities of the UE. Therefore, a new set of signaling needs to be defined to indicate UE capabilities related to CSI prediction and auxiliary information about the UE channel environment to help the network 130 configure an appropriate CSI prediction model for the UE 116.

[0124] As the channel environment of the UE changes, the performance of the currently configured CSI prediction model may degrade over time. Therefore, it is necessary to define a metric to measure the effectiveness of the currently used CSI prediction model and define a set of new signaling to send performance monitoring results from the UE 116 to the network 130.

[0125] If the currently used CSI prediction model is considered invalid, it may be necessary to update, switch or disable the current prediction model. Therefore, a new set of signaling needs to be defined to support such model update, switch and disable operations.

[0126] The present disclosure relates to communication systems.The present disclosure relates to defining functions and processes that support lifecycle management for CSI prediction in cellular systems.

[0127] The present disclosure also relates to auxiliary information indicating UE capabilities related to CSI prediction and channel environment about the UE.

[0128] The present disclosure also relates to defining performance metrics to measure the effectiveness of CSI prediction models, monitoring the performance of the used CSI prediction models, and reporting to the network 130.

[0129] The present disclosure also relates to instructing switching, updating and disabling of used CSI prediction models.

[0130] Embodiments of the present disclosure for lifecycle management of CSI prediction in a cellular system are summarized below and are further fully described herein.

[0131] Method and apparatus for indicating UE capabilities related to CSI prediction and auxiliary information about the channel environment of the UE.

[0132] Methods and apparatus for measuring the effectiveness of a used CSI prediction model and reporting performance monitoring results.

[0133] Methods and apparatus for signaling switching, updating and disabling of used CSI prediction models.

[0134] Fig.16 1 shows a flow chart of an example UE process 1600 for supporting lifecycle management for CSI prediction. For example, the process 1600 for supporting lifecycle management for CSI prediction may be performed by Figure 3UE 116 performs, and the corresponding process can be performed by Figure 1 The example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0135] The process begins at 1610, where the UE sends to the network 130 its capabilities related to CSI prediction, including supported models (AI / ML based or non-AI / ML based), supported neural network types / complexity, and support for model switching, switching, updating, training, and fallback. This signaling can be part of the initial system registration or handover process, which can be in response to a request from the network 130. UECapabilityEnquiry Message. In 1620, parameters and / or models (AI / ML based or non-AI / ML based) for CSI prediction are then provided to the UE from the network 130, for example, via a model ID, a model conversion or a model description, and a feedback format. In 1630, information related to CSI prediction model monitoring is then provided to the UE 116 from the network 130, including performance indicators to be monitored, triggering events for sending reports, and / or any auxiliary information transmitted to the network 130. In 1640, if the triggering event is met and / or there is any auxiliary information, the UE 116 then sends a CSI prediction model monitoring report to the network 130. In 1650, the UE 116 is then provided with information to perform CSI prediction model switching, updating, retraining, or fall back to a default method from the network 130.

[0136] refer to Fig.16 , the discussion illustrates example processes for a UE to provide UE capabilities related to CSI prediction, receive parameters and models for CSI prediction, perform model performance monitoring, and receive CSI prediction model updates, switching, and / or fallback.

[0137] refer to Fig.16 At 1610, the UE 116 capability signaling may include an indication of whether the UE 116 supports CSI prediction. For example, it may be a Boolean indication in the capability signaling.

[0138] UE 116 capability signaling may include an indication of supported methods for future CSI prediction. In one example, UE 116 may indicate that it supports Figure 8 The AI / ML-based CSI prediction method shown. In another example, UE 116 may indicate that it supports non-AI / ML-based CSI prediction methods, such as methods based on extended Kalman filters (EKF). UE 116 may additionally indicate whether model parameters can be updated or reconfigured, for example, to adapt the model to different channel environments, etc. AI / ML-related UE capability signaling details will be disclosed in this disclosure.

[0139] UE 116 capability signaling may include an indication of the maximum time span that UE 116 can predict, i.e., how far into the future UE 116 can predict. This indication may be associated with additional indications of the channel environment of UE 116 (such as UE speed), and / or confidence or reliability of the prediction.

[0140] UE 116 capability signaling may include an indication of the maximum number of predictions that UE 116 can support, i.e., how many future instances UE 116 can predict.

[0141] The UE 116 capability signaling may include an indication of the processing latency associated with the CSI prediction, i.e., the processing time, which may be a number of symbols and / or time slots, required for the UE 116 to perform the prediction and prepare the CSI report message.

[0142] UE 116 capability signaling may include an indication of the confidence and / or reliability of the CSI prediction, i.e., the accuracy of the CSI prediction in terms of a particular measure such as normalized mean square error (NMSE), cosine similarity, etc., which may be provided separately for different channel environments, e.g., LOS / NLOS, UMa / UMi / InH / suburban, different UE speeds, etc.

[0143] If the UE 116 indicates support for AI / ML-based CSI prediction, the UE 116 capability signaling may also include an indication of the AI / ML-related capabilities of the UE 116. In one example, the UE 116 capability signaling may include an indication of supported neural network capabilities. For example, the capability information may include supported neural network types, such as perceptron, feedforward, CNN (convolutional neural network), RNN (recurrent neural network), LSTM (long short-term memory), GAN (generative adversarial network), attention layer, transformer, etc. As another example, the capability information may include the level of supported neural network complexity, such as in terms of the number of layers, neurons, connections, parameters, floating point operations per second (FLOPs), or in terms of model size, for example, in kilobytes, megabytes, etc. As another example, the capability information may include supported neural network activation functions, such as logistic function (sigmoid), hyperbolic function (tanh), and rectified linear unit (ReLU) function, etc.

[0144] In another example, the UE 116 capability information includes an indication of a list of supported AI / ML models for CSI prediction. For example, the UE 116 may have multiple models trained specifically for the site. As an example, it may be a series of Boolean indications for supporting CSI prediction models trained for different channel environments, including but not limited to urban macrocell (Uma) / urban microcell (Umi) / indoor hotspot (InH) / suburban, line of sight (LOS) / non-LOS (NLOS), and a set of different UE speeds, etc.

[0145] In another example, the UE 116 capability information includes an indication of the type of training mode supported. For example, it may indicate whether the UE 116 supports online / offline training and / or transfer learning, etc. For example, the CSI prediction model may be fine-tuned, for example, using transfer learning. In one example, the UE 116 may have a trained CSI prediction model for one channel environment, and retrain the model for a different channel environment by fine-tuning the parameter values ​​of the last output layer from the previously trained prediction model while fixing the structure and parameter values ​​of the remaining layers. In another example, the UE 116 may use a model previously trained for one channel environment as an initialization for performing model retraining for a different channel environment.

[0146] UE 116 capability signaling may include an indication of supported model switching, conversion, or reconfiguration. Model switching capability relates to the ability of UE 116 to switch CSI prediction models between multiple models supported by UE 116. Model switching capability may also include the latency necessary for UE 116 to perform model switching. Model conversion capability relates to the ability of UE 116 to operate a model it receives from network 130 in a compiled format or through a model description. Model reconfiguration capability relates to the ability of UE 116 to update neural network parameters and / or structure from one setting to another. Structural neural network updates may include increasing / decreasing the number of layers, updating connections between layers, changing the order in which layers are placed, enabling / disabling feed-forward skip connections or cascading.

[0147] refer to Fig.16 In 1620, in order to help the network 130 configure an appropriate CSI prediction model, the UE may provide auxiliary information to the network 130, because the model may be configured differently for different environments. In one example, the auxiliary information may include the channel environment of the UE (such as UMa / UMi / InH / suburban), clutter / presence of jams / density / severity, LOS / NLOS indication, indoor / outdoor indication, in-vehicle indication, in-building indication, mobility represented by speed or speed classification, for example, pedestrian / vehicle / high-speed train, etc. The auxiliary information may also include confidence and / or reliability of the prediction model for different channel environments.

[0148] If the UE supports only one CSI prediction model, the network 130 may use a Boolean indication to indicate to the UE 116 to activate / deactivate CSI prediction. If the UE 116 supports more than one CSI prediction model, the network 130 may use a model ID to indicate the selected CSI prediction model to the UE 116. In this case, the multiple models supported by the UE 116 may be registered with the network 130 during the UE capability indication and assigned unique IDs. The deactivation of CSI prediction may be indicated separately using a Boolean indication or by indicating a specific reserved value (e.g., null) in the model ID.

[0149] UE 116 may receive the CSI prediction model transmitted from network 130 or receive the CSI prediction model based on a model description including model structure and / or parameter values ​​received from network 130. The transmitted model or provided model description should remain within the indicated UE capabilities.

[0150] Fig.17 FIG. 1 is a flow chart showing an example UE process 1700 for monitoring CSI prediction performance according to an embodiment of the present disclosure. For example, process 1700 may be performed by Figure 3 UE 116 performs, and the corresponding process can be performed by Figure 1 The example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0151] The process starts at 1710, where a metric for measuring CSI prediction performance, parameters related to filtering the calculated metric, and one or more events for sending a performance monitoring report are provided to the UE from the network 130. In 1720, the UE 116 then calculates the metric value by comparing the previously predicted CSI with the true CSI. In 1730, the UE 116 then filters the calculated metric value and evaluates whether the conditions set for the event for sending the performance monitoring report are met. In 1740, if the event for sending the report is triggered, the UE 116 then sends the performance monitoring report.

[0152] Fig.18 An example timeline 1800 for CSI prediction performance monitoring according to an embodiment of the present disclosure is shown. For example, the timeline 1800 for predictive performance monitoring may be Figure 3 UE 116 utilizes. This example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0153] As an example, the network 130 may be composed of n ref_newThe CSI-RS is sent at an instance represented by n, which is consistent with one of the instances of CSI previously predicted by UE 116 (denoted by n ref_prev +o). This is one way for UE 116 to gain knowledge about the true CSI that it predicted previously. In another example, UE 116 can use a newer CSI reference resource (denoted by n at a later time) at a later time. ref_new ) to predict again its previous use of n ref_prev The future instance n of the CSI reference resource prediction at ref_prev +o, which may be related to the predicted instance n ref_prev +o is inconsistent. This is another way for UE 116 to obtain knowledge about the reference CSI for comparison, although the reference CSI is not the real CSI. In another example, the real CSI can be provided to UE 116 by network 130 for comparison. For example, in a TDD system, network 130 can have the real CSI obtained from the SRS transmitted by the UE and provide the real CSI to UE 116. For example, the performance monitoring report can indicate and / or include at least one of the predicted CSI, the real CSI, the performance indicator, and the timestamp.

[0154] Network 130 may instruct UE 116 to use n for the current instance. ref_new For the current example, UE 116 previously sent a CSI report based on the CSI reference resource at n ref_prev The CSI reference resource at the location reports the predicted CSI, so that the network 130 can be used to calculate the predicted CSI. ref_new The predicted CSI and n ref_new The current real CSI at is compared.

[0155] Alternatively, the network 130 provides a metric (eg, NMSE, cosine similarity, etc.) to the UE 116 to measure the ref_new n ref_prev The previous predicted CSI at n ref_new The network 130 may instruct the UE 116 to perform the CSI calculation with and without n. ref_new For example, the received information related to monitoring the performance of the ML model for predicting CSI may include and / or indicate (i) performance indicators for measuring the performance of the ML model for CSI prediction, including: NMSE, cosine similarity-based metrics, including squared generalized cosine similarity (SGCS), throughput, block error rate (BLER) and / or acknowledgement (ACK) / negative acknowledgement (NACK), and / or (ii) information related to monitoring periodicity.

[0156] Alternatively, the network 130 provides the metric as described above and the threshold to the UE 116, so that if the metric value is greater than or less than the threshold, the UE 116 reports the metric value with and without the current CSI. The threshold can be set for any of the CSI components, including but not limited to CQI, PMI, CRI, SSBRI, LI, RI and / or L1-RSRP. For some metrics, the smaller the value, the smaller the difference between the two input CSIs, for example, NMSE. In this case, if the metric value is greater than a certain threshold, the UE 116 can send a report. For some other metrics, the smaller the metric value, the greater the difference between the two input CSIs, for example, cosine similarity or any variant thereof. In this case, if the metric value is less than a certain threshold, the UE 116 can send a report. For example, the information related to the transmission performance monitoring report received by the UE may include / indicate at least one of the following items: (i) a trigger condition for the transmission performance monitoring report, where the trigger condition is indicated by one or more thresholds of the corresponding one or more performance indicators, and (ii) an uplink channel (e.g., a physical uplink control channel (PUCCH)) for the transmission performance monitoring report.

[0157] The network 130 may instruct the UE 116 to send the current CSI and / or metric value by comparing the previously predicted CSI and the actual CSI. The indication may be provided by the network 130 to the UE 116 via a PDCCH providing a DCI triggering aperiodic CSI report, a medium access control (MAC) control element (CE), or any higher layer signaling.

[0158] The network 130 may indicate to the UE 116 that certain predicted CSI for future instances may be used to compare with the real CSI at the future instances. For periodic CSI, this may be implicitly indicated to the UE 116 from the network 130 via the periodic CSI configuration and the configuration for the predicted future instances. The UE 116 derives instances that are consistent between instances of the CSI reference resource and instances of the CSI prediction. The UE 116 then stores the predicted CSI of such consistent instances for comparison with the real CSI that becomes available in the future. For aperiodic CSI, the network 130 may indicate to the UE 116 via the PDCCH that provides the DCI that triggers the aperiodic CSI report that certain predicted CSI requested by the network 130 may be used to compare with the real CSI in the future. By such an indication, the UE 116 recognizes that the predicted CSI is to be stored for future use. For aperiodic CSI, the network 130 may indicate to the UE 116 via the PDCCH that provides the DCI that triggers the aperiodic CSI report that the network 130 will configure one or more future instances of the CSI reference resource. With such an indication, the UE 116 can derive instances of consistency between future CSI reference resource instances and instances at which the UE 116 is requested to send predicted CSI reports. Therefore, the UE 116 stores the predicted CSI of such consistent instances for future comparison with the actual CSI.

[0159] Fig.19 An example timeline 1900 for measuring and filtering CSI prediction performance according to an embodiment of the present disclosure is shown. For example, the example timeline 1900 for measuring and filtering CSI prediction performance can be composed of Figure 1 Any of the UEs 111 to 116 utilizes. This example is for illustration only, and other embodiments may be used without departing from the scope of the present disclosure.

[0160] The network 130 provides the UE 116 with a metric (e.g., NMSE, cosine similarity, etc.) to measure the difference between the previously predicted CSI and the current true CSI, as well as methods and / or parameters for performing the averaging of these measurements. In one example, for a given instance, the averaging may be performed using the true CSI and one or more previously predicted CSIs for the instance. That is, ,in It is a measurement example n The real CSI with CSI reference resources and the instance n In the example k The measure of the difference between the predicted CSI at is applied to the measured value The weight of the filter. Conditions, such as thresholds, are set for events that trigger performance reporting. In another example, filtering can be performed over time on a series of measurements between the actual CSI of an instance and a set of previously predicted CSIs for the instance. That is, ,in , as described above. If for instance n If there is only one previously predicted CSI, then based on the instance n The measured value of the CSI reference resource at Without weighted summation, it can be simplified to In the above equation, the factor Determines the most recent measurement Compared with the previous average The weight between .

[0161] Other performance indicators for monitoring the performance of the currently used CSI prediction model may be provided from the network 130 to the UE 116, as well as a triggering event for sending a report to the network 130 when the condition is met. The report may also be periodically triggered by a periodicity set by the network 130, or aperiodically triggered by the network 130 via L1 / L2 or any higher layer signaling. As an example, a signal may be sent from the network 130 to the UE to monitor the PDCCH and / or physical downlink shared channel (PDSCH) decoding error rate, and if the measured error rate is greater than or less than an indicated target value or a predefined target value with an optional indicated margin or a predefined margin, or if the measured error rate exceeds an indicated target value range or a predetermined target value range with a maximum / minimum value, a report is sent. The UE 116 may be configured with a window duration in units of time or in units of the number of PDCCH and / or PDSCH receptions to average and measure the error rate. If hybrid automatic repeat request (HARQ) is applied, the error rate of the initial reception or the final reception may be measured. As another example, the UE 116 may be signaled from the network 130 to monitor the received signal-to-noise ratio (SNR) or spectral efficiency and send a report if the measured value is greater than or less than a target value with an optional margin, or if the measured error rate is outside a target value range with a maximum / minimum value. The target value may be predefined, indicated by the network 130, or set by the UE 116 itself based on an estimated effective SNR or spectral efficiency calculated when the UE 116 obtains the CSI. The UE 116 may be configured with a window duration in units of time or in units of the number of measurements (e.g., CSI-RS measurements for target value calculations, PDCCH / PDSCH measurements for received SNR or spectral efficiency calculations).

[0162] refer to Fig.16At 1650, the CSI prediction model ID may be indicated to the UE 116 from the network 130 via L1 / L2 or higher layer signaling to perform model switching. As an example, if the UE 116 has multiple models trained specifically for the site, the switching may be part of the RRC reconfiguration during the handover.

[0163] The UE 116 may be instructed to disable CSI prediction via L1 / L2 or higher layer signaling from the network 130. For example, the UE 116 may receive (e.g., from the network 130 or the BS 102) an indication to switch the ML model used to predict CSI to a non-ML based CSI reporting method, determine a CSI report using the non-ML based CSI reporting method, and transmit a channel (e.g., PUCCH) together with the CSI report.

[0164] A new CSI prediction model may be provided to UE 116 from network 130, for example, via model conversion or a new model description including a neural network structure, parameter values, so that UE 116 may reconfigure the neural network parameter values, or both the neural network structure and the parameter values.

[0165] The network 130 may instruct the UE 116 to perform model retraining or fine-tuning, for example, via transfer learning as described above. The network 130 may provide the UE 116 with a data set or a data set ID so that the UE 116 can download the data set from an external server for training the CSI prediction model. The network 130 may also provide the UE 116 with one or more time offset values ​​of the value of the CSI predicted from the current to the time of training the model. The network 130 may indicate to the UE 116 the cost function to be used during training, such as NMSE, cosine similarity, squared cosine similarity, etc. The network 130 may indicate to the UE 116 whether to perform full model retraining on the entire prediction model or to fine-tune only the final output layer parameter value.

[0166] In an embodiment of the present disclosure, a method for a user equipment (UE) to report channel state information (CSI), the method comprising: receiving: first information related to monitoring the performance of a machine learning (ML) model for predicting CSI, second information related to transmitting a performance monitoring report, third information related to reception of a CSI reference signal (CSI-RS) for determining true CSI on a cell, and a CSI-RS based on the third information; determining: determining the true CSI based on reception of the CSI-RS, and determining a performance monitoring report of the ML model based on the first information and the true CSI; and transmitting a channel with the performance monitoring report based on the second information.

[0167] In an embodiment of the present disclosure, the first information indicates at least one of the following: a performance indicator for measuring the performance of an ML model for CSI prediction, the performance indicator comprising: normalized mean square error (NMSE); a cosine similarity-based metric, including squared generalized cosine similarity (SGCS); throughput; block error rate (BLER), or acknowledgment (ACK) / negative acknowledgment (NACK); and information related to monitoring periodicity.

[0168] In an embodiment of the present disclosure, the second information indicates at least one of: a trigger condition for transmitting a performance monitoring report, wherein the trigger condition is indicated by one or more thresholds regarding the corresponding one or more performance indicators; and an uplink channel for transmitting the performance monitoring report.

[0169] In an embodiment of the present disclosure, the performance monitoring report indicates at least one of the following: predicted CSI, actual CSI, performance indicator, and timestamp.

[0170] In an embodiment of the present disclosure, the determination of the performance monitoring report is based on one or more predicted CSIs for one or more time instances.

[0171] In an embodiment of the present disclosure, the method also includes: receiving a physical downlink control channel (PDCCH), which provides a downlink control information (DCI) format indicating the transmission of a performance monitoring report; and wherein transmitting a channel with a performance monitoring report includes: transmitting a channel with a performance monitoring report based on reception of a PDCCH providing a DCI format.

[0172] In an embodiment of the present disclosure, the method further includes: receiving an indication to switch the ML model for predicting CSI to a non-ML based CSI reporting method; determining a CSI report using the non-ML based CSI reporting method; and transmitting a channel with the CSI report.

[0173] In an embodiment of the present disclosure, a user equipment (UE) includes: a transceiver, configured to: receive first information related to monitoring performance of a machine learning (ML) model for predicting CSI, receive second information related to transmitting a performance monitoring report, receive third information related to reception of a CSI reference signal (CSI-RS) for determining true CSI on a cell, and receive the CSI-RS based on the third information; and a processor, operably connected to the transceiver, configured to: determine the true CSI based on reception of the CSI-RS, and determine a performance monitoring report of the ML model based on the first information and the true CSI, wherein the transceiver is further configured to: transmit a channel with the performance monitoring report based on the second information.

[0174] In an embodiment of the present disclosure, the first information indicates at least one of the following: a performance indicator for measuring the performance of an ML model for CSI prediction, the performance indicator comprising: normalized mean square error (NMSE); a cosine similarity-based metric, including squared generalized cosine similarity (SGCS); throughput; block error rate (BLER), or acknowledgment (ACK) / negative acknowledgment (NACK); and information related to monitoring periodicity.

[0175] In an embodiment of the present disclosure, the second information indicates at least one of: a trigger condition for transmitting a performance monitoring report, wherein the trigger condition is indicated by one or more thresholds regarding the corresponding one or more performance indicators; and an uplink channel for transmitting the performance monitoring report.

[0176] In an embodiment of the present disclosure, the performance monitoring report indicates at least one of the following: predicted CSI, actual CSI, performance indicator, and timestamp.

[0177] In an embodiment of the present disclosure, the determination of the performance monitoring report is based on one or more predicted CSIs for one or more time instances.

[0178] In an embodiment of the present disclosure, the transceiver is further configured to: receive a physical downlink control channel (PDCCH) that provides a downlink control information (DCI) format indicating the transmission of a performance monitoring report; and transmit a channel with a performance monitoring report based on reception of the PDCCH that provides the DCI format.

[0179] In an embodiment of the present disclosure, the transceiver is further configured to: receive an indication to switch the ML model used to predict CSI to a non-ML based CSI reporting method; the processor is further configured to determine a CSI report using the non-ML based CSI reporting method; and the transceiver is further configured to transmit a channel having a CSI report.

[0180] In an embodiment of the present disclosure, a base station includes: a transceiver configured to: transmit first information related to monitoring performance of a machine learning (ML) model for predicting CSI, transmit second information related to transmitting a performance monitoring report, transmit third information related to transmission of a CSI reference signal (CSI-RS) for determining true CSI on a cell, transmit the CSI-RS based on the third information, and receive a channel having a performance monitoring report of the ML model based on the first information and the true CSI based on the second information.

[0181] In an embodiment of the present disclosure, the first information indicates at least one of the following: a performance indicator for measuring the performance of the ML model for CSI prediction, the performance indicator comprising: normalized mean square error (NMSE); a cosine similarity-based metric, including squared generalized cosine similarity (SGCS); throughput; block error rate (BLER), or acknowledgment (ACK) / negative acknowledgment (NACK); and information related to monitoring periodicity.

[0182] In an embodiment of the present disclosure, the second information indicates at least one of: a trigger condition for transmitting a performance monitoring report, wherein the trigger condition is indicated by one or more thresholds regarding the corresponding one or more performance indicators; and an uplink channel for transmitting the performance monitoring report.

[0183] In an embodiment of the present disclosure, the performance monitoring report indicates at least one of the following: predicted CSI, actual CSI, performance indicator, and timestamp.

[0184] In an embodiment of the present disclosure, the performance monitoring report is based on one or more predicted CSIs for one or more time instances.

[0185] In an embodiment of the present disclosure, the transceiver is further configured to: transmit a physical downlink control channel (PDCCH) providing a downlink control information (DCI) format indicating the transmission of a performance monitoring report; and receive a channel with a performance monitoring report based on the transmission of the PDCCH providing the DCI format.

[0186] Fig. 20 The structure of a UE according to an embodiment of the present disclosure is shown.

[0187] like Fig. 20 As shown, the UE according to the embodiment may include a transceiver 2010, a memory 2020, and a processor 2030. The transceiver 2010, the memory 2020, and the processor 2030 of the UE may operate according to the communication method of the above-mentioned UE. However, the components of the UE are not limited thereto. For example, the UE may include more or fewer components than the above-mentioned components. In addition, the processor 2030, the transceiver 2010, and the memory 2020 may be implemented as a single chip. In addition, the processor 2030 may include at least one processor. In addition, Fig. 20 The UEs correspond to Figure 1 UEs 111, 112, 113, 114, 115, and 116.

[0188] The transceiver 2010 generally refers to a UE receiver and a UE transmitter, and can transmit / receive signals to / from a base station or a network entity. The signals transmitted to or received from a base station or a network entity may include control information and data. The transceiver 2010 may include an RF transmitter for up-converting and amplifying the frequency of the transmitted signal, and an RF receiver for low-noise amplification and down-conversion of the frequency of the received signal. However, this is only an example of the transceiver 2010, and the components of the transceiver 2010 are not limited to the RF transmitter and the RF receiver.

[0189] In addition, the transceiver 2010 may receive a signal through a wireless channel and output it to the processor 2030 , and transmit a signal output from the processor 2030 through a wireless channel.

[0190] The memory 2020 may store programs and data required for the operation of the UE. In addition, the memory 2020 may store control information or data included in a signal obtained by the UE. The memory 2020 may be a storage medium such as a read-only memory (ROM), a random access memory (RAM), a hard disk, a CD-ROM, and a DVD, or a combination of storage media.

[0191] The processor 2030 may control a series of processes so that the UE operates as described above. For example, the transceiver 2010 may receive a data signal including a control signal transmitted by a base station or a network entity, and the processor 2030 may determine a result of receiving the control signal and the data signal transmitted by the base station or the network entity.

[0192] Fig.21 The structure of a base station according to an embodiment of the present disclosure is shown.

[0193] like Fig.21 As shown, the base station according to the embodiment may include a transceiver 2110, a memory 2120, and a processor 2130. The transceiver 2110, the memory 2120, and the processor 2130 of the base station may operate according to the communication method of the above-mentioned base station. However, the components of the base station are not limited thereto. For example, the base station may include more or fewer components than the above-mentioned components. In addition, the processor 2130, the transceiver 2110, and the memory 2120 may be implemented as a single chip. In addition, the processor 2130 may include at least one processor. In addition, Fig.21 The base station corresponds to a base station (eg, Figure 1 BS 101, 102, 103).

[0194] The transceiver 2110 generally refers to a base station receiver and a base station transmitter, and can transmit / receive signals to / from a terminal (UE) or a network entity. The signals transmitted to or received from the terminal or network entity may include control information and data. The transceiver 2110 may include an RF transmitter for up-converting and amplifying the frequency of the transmitted signal, and an RF receiver for low-noise amplification and down-conversion of the frequency of the received signal. However, this is only an example of the transceiver 2110, and the components of the transceiver 2110 are not limited to the RF transmitter and the RF receiver.

[0195] In addition, the transceiver 2110 may receive a signal through a wireless channel and output it to the processor 2130 , and transmit a signal output from the processor 2130 through a wireless channel.

[0196] The memory 2120 may store programs and data required for the operation of the base station. In addition, the memory 2120 may store control information or data included in a signal obtained by the base station. The memory 2120 may be a storage medium such as a read-only memory (ROM), a random access memory (RAM), a hard disk, a CD-ROM, and a DVD, or a combination of storage media.

[0197] The processor 2130 may control a series of processes so that the base station operates as described above. For example, the transceiver 2110 may receive a data signal including a control signal transmitted by a terminal, and the processor 2130 may determine a result of receiving the control signal and the data signal transmitted by the terminal.

[0198] Any of the above variations can be used independently or in combination with at least one other variation. The above flow charts show example methods that can be implemented according to the principles of the present disclosure, and various changes can be made to the methods shown in the flow charts herein. For example, although shown as a series of steps, the individual steps in each figure can overlap, occur in parallel, occur in different orders, or occur multiple times. In another example, a step can be omitted or replaced by another step.

[0199] Although the present disclosure has been described with exemplary embodiments, various changes and modifications may be proposed to those skilled in the art. The present disclosure is intended to include such changes and modifications that fall within the scope of the appended claims. Any description in this application should not be construed as implying that any particular element, step or function is an essential element that must be included within the scope of the claims. The scope of the patent subject matter is limited only by the claims.

Claims

1. A method performed by a user equipment UE in a wireless communication system, the method comprising: take over: first information related to monitoring the performance of a machine learning (ML) model for predicting CSI, Second information related to the emission performance monitoring report, third information related to receiving a CSI reference signal CSI-RS for determining true CSI on a cell, and The CSI-RS based on the third information; determining the true CSI based on receiving the CSI-RS, and Determining a performance monitoring report of the ML model based on the first information and the real CSI; as well as Based on the second information, a channel with the performance monitoring report is transmitted.

2. The method according to claim 1, wherein: The first information indicates at least one of the following: A performance indicator for measuring the performance of the ML model for CSI prediction, the performance indicator comprising: Normalized mean square error NMSE, Cosine similarity-based metrics, including squared generalized cosine similarity (SGCS), Throughput, Block Error Rate BLER, or Acknowledgement ACK / Negative Acknowledgement NACK, and Information related to monitoring periodicity.

3. The method according to claim 1, wherein: The second information indicates at least one of the following: a trigger condition for transmitting the performance monitoring report, wherein the trigger condition is indicated by one or more thresholds for corresponding one or more performance indicators, and An uplink channel for transmitting the performance monitoring report.

4. The method according to claim 1, wherein: The performance monitoring report indicates at least one of the following: According to the predicted CSI, The real CSI, performance indicators, and Timestamp.

5. The method according to claim 1, wherein: Determining the performance monitoring report is based on one or more predicted CSIs for one or more time instances.

6. The method according to claim 1, further comprising: receiving a physical downlink control channel PDCCH, the PDCCH providing a downlink control information DCI format indicating transmission of the performance monitoring report; as well as The transmitting the channel having the performance monitoring report comprises: transmitting the channel having the performance monitoring report based on receiving the PDCCH providing the DCI format.

7. The method according to claim 1, further comprising: receiving an indication to switch the ML model for predicting CSI to a non-ML based CSI reporting method; Determining a CSI report using the non-ML based CSI reporting method; as well as A channel with the CSI report is transmitted.

8. A user equipment UE in a wireless communication system, the UE comprising: Transceiver; A controller is connected to the transceiver and is configured to: receiving first information related to monitoring performance of a machine learning (ML) model for predicting CSI, receiving second information related to a transmission performance monitoring report, receiving third information related to receiving a CSI reference signal CSI-RS for determining true CSI on the cell, and Based on the third information, receiving the CSI-RS; as well as determining the true CSI based on receiving the CSI-RS, and determining a performance monitoring report of the ML model based on the first information and the real CSI, The transceiver is further configured to: transmit a channel having the performance monitoring report based on the second information.

9. The UE according to claim 8, wherein: The first information indicates at least one of the following: A performance indicator for measuring the performance of the ML model for CSI prediction, the performance indicator comprising: Normalized mean square error NMSE, Cosine similarity-based metrics, including squared generalized cosine similarity (SGCS), Throughput, Block Error Rate BLER, or Acknowledgement ACK / Negative Acknowledgement NACK, and Information related to monitoring periodicity.

10. The UE according to claim 8, wherein: The second information indicates at least one of the following: a trigger condition for transmitting the performance monitoring report, wherein the trigger condition is indicated by one or more thresholds for corresponding one or more performance indicators, and An uplink channel for transmitting the performance monitoring report.

11. The UE according to claim 8, wherein: The performance monitoring report indicates at least one of the following: According to the predicted CSI, The real CSI, performance indicators, and Timestamp.

12. The UE according to claim 8, wherein: Determining the performance monitoring report is based on one or more predicted CSIs for one or more time instances.

13. The UE according to claim 8, wherein: The controller is also configured to: receiving a physical downlink control channel PDCCH, the PDCCH providing a downlink control information DCI format indicating transmission of the performance monitoring report; and Based on receiving the PDCCH providing the DCI format, the channel with the performance monitoring report is transmitted.

14. A base station in a wireless communication system, the base station comprising: Transceiver; as well as A controller is connected to the transceiver and is configured to: transmitting first information related to monitoring performance of a machine learning (ML) model for predicting CSI, transmitting second information related to the transmission performance monitoring report, transmitting third information related to transmitting a CSI reference signal CSI-RS for determining real CSI on a cell, Based on the third information, transmitting the CSI-RS, and Based on the second information, a channel with a performance monitoring report of the ML model based on the first information and the true CSI is received.

15. A method performed by a base station in a wireless communication system, the method comprising: transmitting first information related to monitoring performance of a machine learning (ML) model for predicting CSI, transmitting second information related to the transmission performance monitoring report, transmitting third information related to transmitting a CSI reference signal CSI-RS for determining real CSI on a cell, Based on the third information, transmitting the CSI-RS, and Based on the second information, a channel with a performance monitoring report of the ML model based on the first information and the true CSI is received.

Images