WO2024122688A1 - Method for performing channel state report in wireless communication system, and apparatus therefor - Google Patents

Abstract

The present specification proposes a method by which a plurality of terminals report channel state information (CSI) in a wireless system. More particularly, the method comprises the steps of: receiving, from a base station, configuration information related to a CSI report, wherein the configuration information comprises (i) common gradient information and/or (ii) weight information about at least one weight for linear combination of at least one information stream related to the CSI; receiving a CSI reference signal (CSI-RS) from the base station; and transmitting the CSI to the base station, wherein the CSI is used as an input for one element related to one terminal from among elements configuring a first layer of a decoder neural network included in the base station.

Description

Method and device for performing channel status reporting in a wireless communication system

This specification relates to a wireless communication system, and more specifically, to a method and device for performing channel state reporting in a wireless communication system.

Wireless communication systems are being widely deployed to provide various types of communication services such as voice and data. In general, a wireless communication system is a multiple access system that can support communication with multiple users by sharing available system resources (bandwidth, transmission power, etc.). Examples of multiple access systems include Code Division Multiple Access (CDMA) systems, Frequency Division Multiple Access (FDMA) systems, Time Division Multiple Access (TDMA) systems, Space Division Multiple Access (SDMA), and Orthogonal Frequency Division Multiple Access (OFDMA) systems. , SC-FDMA (Single Carrier Frequency Division Multiple Access) system, IDMA (Interleave Division Multiple Access) system, etc.

The purpose of this specification is to provide a method and device for performing channel status reporting in a wireless communication system.

Additionally, the purpose of this specification is to provide a method and device for performing channel state reporting based on a structure suitable for a multi-stream multi-user channel state information (CSI) network in a wireless communication system.

Additionally, the purpose of this specification is to provide a signaling method and apparatus for reducing signaling overhead in a multi-stream multi-user channel state information (CSI) network in a wireless communication system.

The technical problems to be achieved in this specification are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

This specification provides a method and device for performing channel state reporting in a wireless communication system.

More specifically, in this specification, in a method for a plurality of terminals to report channel state information (CSI) in a wireless system, the method performed by one terminal of the plurality of terminals is provided by a base station. , receiving setting information related to CSI reporting, wherein the setting information includes (i) common gradient information and/or (ii) linear combination of at least one information stream related to the CSI. Contains weight information for at least one weight; Receiving a channel state information reference signal (CSI-RS) from the base station; Transmitting the CSI to the base station, wherein the CSI is one of the elements constituting the first layer of the decoder neural network included in the base station, related to the one terminal. It is characterized by being used as input to an element.

In addition, the present specification may be characterized in that the CSI input to the one element is output as an output value generated based on a linear combination of at least one information stream related to the CSI.

In addition, this specification provides that the output value generated based on the linear combination of at least one information stream related to the CSI is input to all or part of the elements constituting the second layer of the decoder neural network included in the base station. It can be characterized as being used.

Additionally, in this specification, the elements constituting the first layer of the decoder neural network and the elements constituting the second layer of the decoder neural network receive vector variables as input, and perform a linear combination of the input vector variables. It may be characterized as a vector neuron that performs operations.

Additionally, the present specification may be characterized in that no activation function is applied between the first layer of the decoder neural network and the second layer of the decoder neural network.

Additionally, the present specification may be characterized in that each of the at least one weight is related to one of the at least one information stream.

Additionally, in this specification, the weight information may be characterized as information about the relative ratios of the remaining weights to a specific weight among the at least one weight.

Additionally, in this specification, the weight information may be characterized as information about the remaining weights excluding information about a specific weight among the at least one weight.

Additionally, the present specification may be characterized in that information about a certain weight is determined based on the weight information under predefined limiting conditions.

In addition, in this specification, the common gradient information is a loss function for the output of a specific element related to a specific terminal among the plurality of terminals among the elements constituting the first layer of the decoder neural network. It may be characterized as information about a gradient vector.

In addition, in this specification, in a wireless communication system, one terminal that reports channel status together with a plurality of terminals includes: a transmitter for transmitting a wireless signal; a receiver for receiving a wireless signal; at least one processor; and at least one computer memory operably connectable to the at least one processor and storing instructions that, when executed by the at least one processor, perform operations, the operations comprising: , receiving setting information related to CSI reporting, wherein the setting information includes (i) common gradient information and/or (ii) linear combination of at least one information stream related to the CSI. Contains weight information for at least one weight; Receiving a channel state information reference signal (CSI-RS) from the base station; Transmitting the CSI to the base station, wherein the CSI is one of the elements constituting the first layer of the decoder neural network included in the base station, related to the one terminal. A terminal characterized in that it is used as an input to an element.

In addition, the present specification provides a method for a base station to receive channel state information from a plurality of terminals in a wireless communication system, including transmitting configuration information related to CSI reporting to each of the plurality of terminals, the configuration information being ( Contains i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI; Transmitting a channel state information reference signal (CSI-RS) to each of the plurality of terminals; Receiving the CSI from each of the plurality of terminals, wherein the CSI is connected to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station. It is characterized by being used as an input for one related element.

Additionally, in this specification, a base station that receives channel state information from a plurality of terminals in a wireless communication system includes: a transmitter for transmitting a wireless signal; A receiver for receiving wireless signals; at least one processor; and at least one computer memory operably connectable to the at least one processor and storing instructions that, when executed by the at least one processor, perform operations, the operations comprising: Transmitting configuration information related to CSI reporting to each of the terminals, wherein the configuration information includes (i) common gradient information and/or (ii) at least one information stream related to the CSI. Contains weight information about at least one weight for linear combination; Transmitting a channel state information reference signal (CSI-RS) to each of the plurality of terminals; Receiving the CSI from each of the plurality of terminals, wherein the CSI is connected to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station. It is characterized by being used as an input for one related element.

In addition, the present specification provides that, in a non-transitory computer readable medium (CRM) storing one or more instructions, one or more instructions executable by one or more processors are provided by a terminal: from a base station, related to CSI reporting. Receive configuration information, wherein the configuration information includes (i) common gradient information and/or (ii) at least one weight for linear combination of at least one information stream related to the CSI. Contains weight information for; Receive a channel state information reference signal (CSI-RS) from the base station; The CSI is transmitted to the base station, and the CSI is for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station. It is characterized by being used as input.

In addition, the present specification provides a device including one or more memories and one or more processors functionally connected to the one or more memories, wherein the one or more processors receive, from a base station, configuration information related to CSI reporting. to receive, and the setting information includes (i) common gradient information and/or (ii) a weight for at least one weight for linear combination of at least one information stream related to the CSI. Contains information; Receive a channel state information reference signal (CSI-RS) from the base station; The CSI is transmitted to the base station, and the CSI is for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station. A device characterized by being used as an input.

This specification has the effect of performing channel state reporting in a wireless communication system.

In addition, this specification has the effect of reducing signaling overhead even if the number of feedback streams increases through channel state reporting based on a structure suitable for a multi-stream multi-user channel state information (CSI) network in a wireless communication system.

The effects that can be obtained in this specification are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. .

The drawings attached below are intended to aid understanding of the present specification and may provide embodiments of the present specification along with a detailed description. However, the technical features of this specification are not limited to specific drawings, and the features disclosed in each drawing may be combined to form a new embodiment. Reference numerals in each drawing may refer to structural elements.

Figure 1 is a diagram showing an example of a communication system applicable to this specification.

Figure 2 is a diagram showing an example of a wireless device applicable to this specification.

Figure 3 is a diagram showing a method of processing a transmission signal applicable to this specification.

Figure 4 is a diagram showing another example of a wireless device applicable to this specification.

Figure 5 is a diagram showing an example of a portable device applicable to this specification.

Figure 6 is a diagram showing physical channels applicable to this specification and a signal transmission method using them.

Figure 7 is a diagram showing the structure of a wireless frame applicable to this specification.

Figure 8 is a diagram showing a slot structure applicable to this specification.

Figure 9 is a diagram showing an example of a communication structure that can be provided in a 6G system applicable to this specification.

Figure 10 is a diagram showing an example of a CSI network structure.

Figures 11 and 12 are diagrams showing an example of a CSI network structure.

Figure 13 is a diagram showing an example of a decoder structure.

Figure 14 is a diagram showing an example of a multiuser multi-stream CSI network.

15 and 16 show backpropagation when the number S of feedback streams from each UE is 2 in a multi-stream CSI network, compared to the case where only a single feedback stream is supported from each UE (S = 1). This is a diagram to explain the problem of doubling the signaling overhead for .

Figure 17 is a diagram showing an example of a multi-user multi-stream CSI network structure proposed in this specification.

Figure 18 is a diagram showing an example of an online learning method to which the decoder head structure proposed in this specification is applied.

Figures 19 and 20 are diagrams showing the effect of the decoder structure proposed in this specification.

Figure 21 is a flowchart showing an example of how the channel state reporting method proposed in this specification is performed in a terminal.

Figure 22 is a flowchart showing an example of how the channel state reporting method proposed in this specification is performed at the base station.

The following embodiments combine the elements and features of the present specification in a predetermined form. Each component or feature may be considered optional unless explicitly stated otherwise. Each component or feature may be implemented in a form that is not combined with other components or features. Additionally, embodiments of the present specification may be configured by combining some components and/or features. The order of operations described in the embodiments of this specification may be changed. Some features or features of one embodiment may be included in another embodiment or may be replaced with corresponding features or features of another embodiment.

In the description of the drawings, procedures or steps that may obscure the gist of the present specification are not described, and procedures or steps that can be understood at the level of those skilled in the art are not described.

Throughout the specification, when a part is said to “comprise or include” a certain element, this means that it does not exclude other elements but may further include other elements, unless specifically stated to the contrary. do. In addition, terms such as "... unit", "... unit", and "module" used in the specification refer to a unit that processes at least one function or operation, which refers to hardware, software, or a combination of hardware and software. It can be implemented as: Additionally, “a or an,” “one,” “the,” and similar related terms are used in the context of describing the present specification (particularly in the context of the following claims) as used herein. It may be used in both singular and plural terms, unless indicated otherwise or clearly contradicted by context.

Embodiments of the present specification have been described focusing on the data transmission and reception relationship between the base station and the mobile station. Here, the base station is meant as a terminal node of the network that directly communicates with the mobile station. Certain operations described herein as being performed by the base station may, in some cases, be performed by an upper node of the base station.

That is, in a network comprised of a plurality of network nodes including a base station, various operations performed for communication with a mobile station may be performed by the base station or other network nodes other than the base station. At this time, 'base station' refers to terms such as fixed station, Node B, eNB (eNode B), gNB (gNode B), ng-eNB, advanced base station (ABS), or access point. It can be replaced by .

In addition, in the embodiments of the present specification, the terminal includes a user equipment (UE), a mobile station (MS), a subscriber station (SS), a mobile subscriber station (MSS), It can be replaced with terms such as mobile terminal or advanced mobile station (AMS).

Additionally, the transmitting end refers to a fixed and/or mobile node that provides a data service or a voice service, and the receiving end refers to a fixed and/or mobile node that receives a data service or a voice service. Therefore, in the case of uplink, the mobile station can be the transmitting end and the base station can be the receiving end. Likewise, in the case of downlink, the mobile station can be the receiving end and the base station can be the transmitting end.

Embodiments of the present specification include wireless access systems such as the IEEE 802.xx system, 3GPP (3rd Generation Partnership Project) system, 3GPP LTE (Long Term Evolution) system, 3GPP 5G (5th generation) NR (New Radio) system, and 3GPP2 system. It may be supported by at least one standard document disclosed in, and in particular, the embodiments of the present specification are supported by the 3GPP TS (technical specification) 38.211, 3GPP TS 38.212, 3GPP TS 38.213, 3GPP TS 38.321 and 3GPP TS 38.331 documents. It can be.

Additionally, embodiments of the present specification can be applied to other wireless access systems and are not limited to the above-described system. As an example, it may be applicable to systems applied after the 3GPP 5G NR system and is not limited to a specific system.

That is, obvious steps or parts that are not described among the embodiments of the present specification can be explained with reference to the documents. Additionally, all terms disclosed in this specification can be explained by the standard document.

Hereinafter, preferred embodiments according to the present specification will be described in detail with reference to the attached drawings. The detailed description to be disclosed below along with the accompanying drawings is intended to describe exemplary embodiments of the present specification and is not intended to represent the only embodiments in which the technical features of the present specification may be implemented.

In addition, specific terms used in the embodiments of the present specification are provided to aid understanding of the present specification, and the use of such specific terms may be changed to other forms without departing from the technical spirit of the present specification.

The following technologies include code division multiple access (CDMA), frequency division multiple access (FDMA), time division multiple access (TDMA), orthogonal frequency division multiple access (OFDMA), and single carrier frequency division multiple access (SC-FDMA). It can be applied to various wireless access systems.

In the following, for clarity of explanation, the description is based on the 3GPP communication system (e.g., LTE, NR, etc.), but the technical idea of the present invention is not limited thereto. LTE is 3GPP TS 36.xxx Release 8 and later. In detail, LTE technology after 3GPP TS 36.xxx Release 10 may be referred to as LTE-A, and LTE technology after 3GPP TS 36.xxx Release 13 may be referred to as LTE-A pro. 3GPP NR may mean technology after TS 38.xxx Release 15, and “xxx” may mean technology after TS Release 17 and/or Release 18. LTE/NR/6G can be collectively referred to as a 3GPP system.

Regarding background technology, terms, abbreviations, etc. used in this specification, reference may be made to matters described in standard documents published prior to the present invention. As an example, you can refer to the 36.xxx and 38.xxx standard documents.

Communication systems applicable to this specification

Although not limited thereto, various descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed herein may be applied to various fields requiring wireless communication/connection (e.g., 5G) between devices.

Hereinafter, a more detailed example will be provided with reference to the drawings. In the following drawings/descriptions, identical reference numerals may illustrate identical or corresponding hardware blocks, software blocks, or functional blocks, unless otherwise noted.

FIG. 1 is a diagram illustrating an example of a communication system applied to the present specification. Referring to FIG. 1, the communication system 100 applied to the present specification includes a wireless device, a base station, and a network. Here, a wireless device refers to a device that performs communication using wireless access technology (e.g., 5G NR, LTE) and may be referred to as a communication/wireless/5G device. Although not limited thereto, wireless devices include robots (100a), vehicles (100b-1, 100b-2), extended reality (XR) devices (100c), hand-held devices (100d), and home appliances (100d). appliance) (100e), IoT (Internet of Thing) device (100f), and AI (artificial intelligence) device/server (100g). For example, vehicles may include vehicles equipped with wireless communication functions, autonomous vehicles, vehicles capable of inter-vehicle communication, etc. Here, the vehicles 100b-1 and 100b-2 may include an unmanned aerial vehicle (UAV) (eg, a drone). The XR device 100c includes augmented reality (AR)/virtual reality (VR)/mixed reality (MR) devices, including a head-mounted device (HMD), a head-up display (HUD) installed in a vehicle, a television, It can be implemented in the form of smartphones, computers, wearable devices, home appliances, digital signage, vehicles, robots, etc. The mobile device 100d may include a smartphone, smart pad, wearable device (eg, smart watch, smart glasses), computer (eg, laptop, etc.), etc. Home appliances 100e may include a TV, refrigerator, washing machine, etc. IoT device 100f may include sensors, smart meters, etc. For example, the base station 120 and the network 130 may also be implemented as wireless devices, and a specific wireless device 120a may operate as a base station/network node for other wireless devices.

Wireless devices 100a to 100f may be connected to the network 130 through the base station 120. AI technology may be applied to the wireless devices 100a to 100f, and the wireless devices 100a to 100f may be connected to the AI server 100g through the network 130. The network 130 may be configured using a 3G network, 4G (eg, LTE) network, or 5G (eg, NR) network. Wireless devices 100a to 100f may communicate with each other through the base station 120/network 130, but communicate directly (e.g., sidelink communication) without going through the base station 120/network 130. You may. For example, vehicles 100b-1 and 100b-2 may communicate directly (eg, vehicle to vehicle (V2V)/vehicle to everything (V2X) communication). Additionally, the IoT device 100f (eg, sensor) may communicate directly with other IoT devices (eg, sensor) or other wireless devices 100a to 100f.

Wireless communication/connection (150a, 150b, 150c) may be established between wireless devices (100a to 100f)/base station (120) and base station (120)/base station (120). Here, wireless communication/connection includes various methods such as uplink/downlink communication (150a), sidelink communication (150b) (or D2D communication), and inter-base station communication (150c) (e.g., relay, integrated access backhaul (IAB)). This can be achieved through wireless access technology (e.g. 5G NR). Through wireless communication/connection (150a, 150b, 150c), a wireless device and a base station/wireless device, and a base station and a base station can transmit/receive wireless signals to each other. For example, wireless communication/ connection 150a, 150b, and 150c may transmit/receive signals through various physical channels. To this end, based on the various proposals in this specification, various configuration information setting processes for transmitting/receiving wireless signals, various signal processing processes (e.g., channel encoding/decoding, modulation/demodulation, resource mapping/demapping, etc.) , at least some of the resource allocation process, etc. may be performed.

Communication systems applicable to this specification

Figure 2 is a diagram showing an example of a wireless device that can be applied to this specification.

Referring to FIG. 2, the first wireless device 200a and the second wireless device 200b can transmit and receive wireless signals through various wireless access technologies (eg, LTE, NR). Here, {first wireless device 200a, second wireless device 200b} refers to {wireless device 100x, base station 120} and/or {wireless device 100x, wireless device 100x) in FIG. } can be responded to.

The first wireless device 200a includes one or more processors 202a and one or more memories 204a, and may further include one or more transceivers 206a and/or one or more antennas 208a. Processor 202a controls memory 204a and/or transceiver 206a and may be configured to implement the descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed herein. For example, the processor 202a may process information in the memory 204a to generate first information/signal and then transmit a wireless signal including the first information/signal through the transceiver 206a. Additionally, the processor 202a may receive a wireless signal including the second information/signal through the transceiver 206a and then store information obtained from signal processing of the second information/signal in the memory 204a. The memory 204a may be connected to the processor 202a and may store various information related to the operation of the processor 202a. For example, memory 204a may perform some or all of the processes controlled by processor 202a or instructions for performing the descriptions, functions, procedures, suggestions, methods, and/or operational flowcharts disclosed herein. Software code containing them can be stored. Here, the processor 202a and the memory 204a may be part of a communication modem/circuit/chip designed to implement wireless communication technology (eg, LTE, NR). Transceiver 206a may be coupled to processor 202a and may transmit and/or receive wireless signals via one or more antennas 208a. Transceiver 206a may include a transmitter and/or receiver. The transceiver 206a may be used interchangeably with a radio frequency (RF) unit. In this specification, a wireless device may mean a communication modem/circuit/chip.

The second wireless device 200b includes one or more processors 202b, one or more memories 204b, and may further include one or more transceivers 206b and/or one or more antennas 208b. Processor 202b controls memory 204b and/or transceiver 206b and may be configured to implement the descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed herein. For example, the processor 202b may process information in the memory 204b to generate third information/signal and then transmit a wireless signal including the third information/signal through the transceiver 206b. Additionally, the processor 202b may receive a wireless signal including the fourth information/signal through the transceiver 206b and then store information obtained from signal processing of the fourth information/signal in the memory 204b. The memory 204b may be connected to the processor 202b and may store various information related to the operation of the processor 202b. For example, memory 204b may perform some or all of the processes controlled by processor 202b or instructions for performing the descriptions, functions, procedures, suggestions, methods, and/or operational flowcharts disclosed herein. Software code containing them can be stored. Here, the processor 202b and the memory 204b may be part of a communication modem/circuit/chip designed to implement wireless communication technology (eg, LTE, NR). Transceiver 206b may be coupled to processor 202b and may transmit and/or receive wireless signals via one or more antennas 208b. The transceiver 206b may include a transmitter and/or a receiver. The transceiver 206b may be used interchangeably with an RF unit. In this specification, a wireless device may mean a communication modem/circuit/chip.

Hereinafter, the hardware elements of the wireless devices 200a and 200b will be described in more detail. Although not limited thereto, one or more protocol layers may be implemented by one or more processors 202a and 202b. For example, one or more processors 202a and 202b may operate on one or more layers (e.g., physical (PHY), media access control (MAC), radio link control (RLC), packet data convergence protocol (PDCP), and radio resource (RRC). control) and functional layers such as SDAP (service data adaptation protocol) can be implemented. One or more processors 202a, 202b may generate one or more Protocol Data Units (PDUs) and/or one or more service data units (SDUs) according to the descriptions, functions, procedures, suggestions, methods, and/or operational flowcharts disclosed herein. can be created. One or more processors 202a and 202b may generate messages, control information, data or information according to the descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed herein. One or more processors 202a and 202b generate signals (e.g., baseband signals) containing PDUs, SDUs, messages, control information, data or information according to the functions, procedures, proposals and/or methods disclosed herein. , can be provided to one or more transceivers (206a, 206b). One or more processors 202a, 202b may receive signals (e.g., baseband signals) from one or more transceivers 206a, 206b, and the descriptions, functions, procedures, suggestions, methods, and/or operational flowcharts disclosed herein. Depending on the device, PDU, SDU, message, control information, data or information can be obtained.

One or more processors 202a, 202b may be referred to as a controller, microcontroller, microprocessor, or microcomputer. One or more processors 202a and 202b may be implemented by hardware, firmware, software, or a combination thereof. As an example, one or more application specific integrated circuits (ASICs), one or more digital signal processors (DSPs), one or more digital signal processing devices (DSPDs), one or more programmable logic devices (PLDs), or one or more field programmable gate arrays (FPGAs) May be included in one or more processors 202a and 202b. The descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed in this specification may be implemented using firmware or software, and the firmware or software may be implemented to include modules, procedures, functions, etc. Firmware or software configured to perform the descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed herein may be included in one or more processors 202a and 202b or stored in one or more memories 204a and 204b. It may be driven by the above processors 202a and 202b. The descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed herein may be implemented using firmware or software in the form of codes, instructions and/or sets of instructions.

One or more memories 204a and 204b may be connected to one or more processors 202a and 202b and may store various types of data, signals, messages, information, programs, codes, instructions and/or commands. One or more memories 204a, 204b may include read only memory (ROM), random access memory (RAM), erasable programmable read only memory (EPROM), flash memory, hard drives, registers, cache memory, computer readable storage media, and/or It may be composed of a combination of these. One or more memories 204a and 204b may be located internal to and/or external to one or more processors 202a and 202b. Additionally, one or more memories 204a and 204b may be connected to one or more processors 202a and 202b through various technologies, such as wired or wireless connections.

One or more transceivers (206a, 206b) may transmit user data, control information, wireless signals/channels, etc. mentioned in the methods and/or operation flowcharts of this specification to one or more other devices. One or more transceivers 206a, 206b may receive user data, control information, wireless signals/channels, etc. referred to in the descriptions, functions, procedures, suggestions, methods and/or operational flowcharts disclosed herein, etc. from one or more other devices. there is. For example, one or more transceivers 206a and 206b may be connected to one or more processors 202a and 202b and may transmit and receive wireless signals. For example, one or more processors 202a and 202b may control one or more transceivers 206a and 206b to transmit user data, control information, or wireless signals to one or more other devices. Additionally, one or more processors 202a and 202b may control one or more transceivers 206a and 206b to receive user data, control information, or wireless signals from one or more other devices. In addition, one or more transceivers (206a, 206b) may be connected to one or more antennas (208a, 208b), and one or more transceivers (206a, 206b) may be connected to the description and functions disclosed herein through one or more antennas (208a, 208b). , may be set to transmit and receive user data, control information, wireless signals/channels, etc. mentioned in procedures, proposals, methods and/or operation flow charts, etc. In this specification, one or more antennas may be a plurality of physical antennas or a plurality of logical antennas (eg, antenna ports). One or more transceivers (206a, 206b) process the received user data, control information, wireless signals/channels, etc. using one or more processors (202a, 202b), and convert the received wireless signals/channels, etc. from the RF band signal. It can be converted to a baseband signal. One or more transceivers (206a, 206b) may convert user data, control information, wireless signals/channels, etc. processed using one or more processors (202a, 202b) from a baseband signal to an RF band signal. To this end, one or more transceivers 206a, 206b may include (analog) oscillators and/or filters.

Figure 3 is a diagram illustrating a method of processing a transmission signal applied to this specification. As an example, the transmission signal may be processed by a signal processing circuit. At this time, the signal processing circuit 300 may include a scrambler 310, a modulator 320, a layer mapper 330, a precoder 340, a resource mapper 350, and a signal generator 360. At this time, as an example, the operation/function of FIG. 3 may be performed in the processors 202a and 202b and/or transceivers 206a and 206b of FIG. 2. Additionally, as an example, the hardware elements of FIG. 3 may be implemented in the processors 202a and 202b and/or transceivers 206a and 206b of FIG. 2. As an example, blocks 310 to 350 may be implemented in the processors 202a and 202b of FIG. 2, and block 360 may be implemented in the transceivers 206a and 206b of FIG. 2, but are not limited to the above-described embodiment.

The codeword can be converted into a wireless signal through the signal processing circuit 300 of FIG. 3. Here, a codeword is an encoded bit sequence of an information block. The information block may include a transport block (eg, UL-SCH transport block, DL-SCH transport block). Wireless signals may be transmitted through various physical channels (eg, PUSCH, PDSCH) in FIG. 6. Specifically, the codeword may be converted into a scrambled bit sequence by the scrambler 310. The scramble sequence used for scrambling is generated based on an initialization value, and the initialization value may include ID information of the wireless device. The scrambled bit sequence may be modulated into a modulation symbol sequence by the modulator 320. Modulation methods may include pi/2-BPSK (pi/2-binary phase shift keying), m-PSK (m-phase shift keying), m-QAM (m-quadrature amplitude modulation), etc.

The complex modulation symbol sequence may be mapped to one or more transport layers by the layer mapper 330. The modulation symbols of each transport layer may be mapped to corresponding antenna port(s) by the precoder 340 (precoding). The output z of the precoder 340 can be obtained by multiplying the output y of the layer mapper 330 by the N*M precoding matrix W. Here, N is the number of antenna ports and M is the number of transport layers. Here, the precoder 340 may perform precoding after performing transform precoding (eg, discrete Fourier transform (DFT) transform) on complex modulation symbols. Additionally, the precoder 340 may perform precoding without performing transform precoding.

The resource mapper 350 can map the modulation symbols of each antenna port to time-frequency resources. A time-frequency resource may include a plurality of symbols (eg, CP-OFDMA symbol, DFT-s-OFDMA symbol) in the time domain and a plurality of subcarriers in the frequency domain. The signal generator 360 generates a wireless signal from the mapped modulation symbols, and the generated wireless signal can be transmitted to another device through each antenna. To this end, the signal generator 360 may include an inverse fast fourier transform (IFFT) module, a cyclic prefix (CP) inserter, a digital-to-analog converter (DAC), a frequency uplink converter, etc. .

The signal processing process for a received signal in a wireless device may be configured as the reverse of the signal processing processes 310 to 360 of FIG. 3. As an example, a wireless device (eg, 200a and 200b in FIG. 2) may receive a wireless signal from the outside through an antenna port/transceiver. The received wireless signal can be converted into a baseband signal through a signal restorer. To this end, the signal restorer may include a frequency downlink converter, an analog-to-digital converter (ADC), a CP remover, and a fast fourier transform (FFT) module. Afterwards, the baseband signal can be restored to a codeword through a resource de-mapper process, postcoding process, demodulation process, and de-scramble process. The codeword can be restored to the original information block through decoding. Accordingly, a signal processing circuit (not shown) for a received signal may include a signal restorer, resource de-mapper, postcoder, demodulator, de-scrambler, and decoder.

Wireless device structure applicable to this specification

Figure 4 is a diagram showing another example of a wireless device applied to this specification.

Referring to FIG. 4, the wireless device 400 corresponds to the wireless devices 200a and 200b of FIG. 2 and includes various elements, components, units/units, and/or modules. ) can be composed of. For example, the wireless device 400 may include a communication unit 410, a control unit 420, a memory unit 430, and an additional element 440. The communication unit may include communication circuitry 412 and transceiver(s) 414. For example, communication circuitry 412 may include one or more processors 202a and 202b and/or one or more memories 204a and 204b of FIG. 2 . For example, transceiver(s) 414 may include one or more transceivers 206a, 206b and/or one or more antennas 208a, 208b of FIG. 2. The control unit 420 is electrically connected to the communication unit 410, the memory unit 430, and the additional element 440 and controls overall operations of the wireless device. For example, the control unit 420 may control the electrical/mechanical operation of the wireless device based on the program/code/command/information stored in the memory unit 430. In addition, the control unit 420 transmits the information stored in the memory unit 430 to the outside (e.g., another communication device) through the communication unit 410 through a wireless/wired interface, or to the outside (e.g., to another communication device) through the communication unit 410. Information received through a wireless/wired interface from another communication device can be stored in the memory unit 430.

The additional element 440 may be configured in various ways depending on the type of wireless device. For example, the additional element 440 may include at least one of a power unit/battery, an input/output unit, a driving unit, and a computing unit. Although not limited thereto, the wireless device 400 may include a robot (FIG. 1, 100a), a vehicle (FIG. 1, 100b-1, 100b-2), an XR device (FIG. 1, 100c), and a portable device (FIG. 1, 100d). ), home appliances (Figure 1, 100e), IoT devices (Figure 1, 100f), digital broadcasting terminals, hologram devices, public safety devices, MTC devices, medical devices, fintech devices (or financial devices), security devices, climate/ It can be implemented in the form of an environmental device, AI server/device (FIG. 1, 140), base station (FIG. 1, 120), network node, etc. Wireless devices can be mobile or used in fixed locations depending on the usage/service.

In FIG. 4 , various elements, components, units/parts, and/or modules within the wireless device 400 may be entirely interconnected through a wired interface, or at least some of them may be wirelessly connected through the communication unit 410. For example, within the wireless device 400, the control unit 420 and the communication unit 410 are connected by wire, and the control unit 420 and the first unit (e.g., 430, 440) are connected wirelessly through the communication unit 410. can be connected Additionally, each element, component, unit/part, and/or module within the wireless device 400 may further include one or more elements. For example, the control unit 420 may be comprised of one or more processor sets. For example, the control unit 420 may be comprised of a communication control processor, an application processor, an electronic control unit (ECU), a graphics processing processor, and a memory control processor. As another example, the memory unit 430 may be comprised of RAM, dynamic RAM (DRAM), ROM, flash memory, volatile memory, non-volatile memory, and/or a combination thereof. It can be configured.

Mobile devices to which this specification applies

Figure 5 is a diagram showing an example of a portable device applied to this specification.

Figure 5 illustrates a portable device to which this specification applies. Portable devices may include smartphones, smart pads, wearable devices (e.g., smart watches, smart glasses), and portable computers (e.g., laptops, etc.). A mobile device may be referred to as a mobile station (MS), user terminal (UT), mobile subscriber station (MSS), subscriber station (SS), advanced mobile station (AMS), or wireless terminal (WT).

Referring to FIG. 5, the portable device 500 includes an antenna unit 508, a communication unit 510, a control unit 520, a memory unit 530, a power supply unit 540a, an interface unit 540b, and an input/output unit 540c. ) may include. The antenna unit 508 may be configured as part of the communication unit 510. Blocks 510 to 530/540a to 540c correspond to blocks 410 to 430/440 in FIG. 4, respectively.

The communication unit 510 can transmit and receive signals (eg, data, control signals, etc.) with other wireless devices and base stations. The control unit 520 can control the components of the portable device 500 to perform various operations. The control unit 520 may include an application processor (AP). The memory unit 530 may store data/parameters/programs/codes/commands necessary for driving the portable device 500. Additionally, the memory unit 530 can store input/output data/information, etc. The power supply unit 540a supplies power to the portable device 500 and may include a wired/wireless charging circuit, a battery, etc. The interface unit 540b may support connection between the mobile device 500 and other external devices. The interface unit 540b may include various ports (eg, audio input/output ports, video input/output ports) for connection to external devices. The input/output unit 540c may input or output video information/signals, audio information/signals, data, and/or information input from the user. The input/output unit 540c may include a camera, a microphone, a user input unit, a display unit 540d, a speaker, and/or a haptic module.

For example, in the case of data communication, the input/output unit 540c acquires information/signals (e.g., touch, text, voice, image, video) input from the user, and the obtained information/signals are stored in the memory unit 530. It can be saved. The communication unit 510 may convert the information/signal stored in the memory into a wireless signal and transmit the converted wireless signal directly to another wireless device or to a base station. Additionally, the communication unit 510 may receive a wireless signal from another wireless device or a base station and then restore the received wireless signal to the original information/signal. The restored information/signal may be stored in the memory unit 530 and then output in various forms (eg, text, voice, image, video, haptic) through the input/output unit 540c.

Physical channels and general signal transmission

In a wireless access system, a terminal can receive information from a base station through downlink (DL) and transmit information to the base station through uplink (UL). Information transmitted and received between the base station and the terminal includes general data information and various control information, and various physical channels exist depending on the type/purpose of the information they transmit and receive.

Figure 6 is a diagram showing physical channels applied to this specification and a signal transmission method using them.

A terminal that is turned on again from a power-off state or newly entered a cell performs an initial cell search task such as synchronizing with the base station in step S611. To this end, the terminal receives the primary synchronization channel (P-SCH) and secondary synchronization channel (S-SCH) from the base station to synchronize with the base station and obtain information such as cell ID. .

Afterwards, the terminal can obtain intra-cell broadcast information by receiving a physical broadcast channel (PBCH) signal from the base station. Meanwhile, the terminal can check the downlink channel status by receiving a downlink reference signal (DL RS) in the initial cell search stage. After completing the initial cell search, the UE receives a physical downlink control channel (PDCCH) and a physical downlink shared channel (PDSCH) according to the physical downlink control channel information in step S612 and further You can obtain specific system information.

Thereafter, the terminal may perform a random access procedure such as steps S613 to S616 to complete access to the base station. To this end, the terminal transmits a preamble through a physical random access channel (PRACH) (S613), and RAR (RAR) for the preamble through the physical downlink control channel and the corresponding physical downlink shared channel. A random access response) can be received (S614). The terminal transmits a physical uplink shared channel (PUSCH) using scheduling information in the RAR (S615), and a contention resolution procedure such as reception of a physical downlink control channel signal and a corresponding physical downlink shared channel signal. ) can be performed (S616).

The terminal that has performed the above-described procedure then receives a physical downlink control channel signal and/or a physical downlink shared channel signal (S617) and a physical uplink shared channel as a general uplink/downlink signal transmission procedure. Transmission of a channel (PUSCH) signal and/or a physical uplink control channel (PUCCH) signal may be performed (S618).

The control information transmitted from the terminal to the base station is collectively referred to as uplink control information (UCI). UCI includes HARQ-ACK/NACK (hybrid automatic repeat and request acknowledgment/negative-ACK), SR (scheduling request), CQI (channel quality indication), PMI (precoding matrix indication), RI (rank indication), and BI (beam indication). ) information, etc. At this time, UCI is generally transmitted periodically through PUCCH, but depending on the embodiment (e.g., when control information and traffic data must be transmitted simultaneously), it may be transmitted through PUSCH. Additionally, at the request/instruction of the network, the terminal may transmit UCI aperiodically through PUSCH.

Figure 7 is a diagram showing the structure of a wireless frame applicable to this specification.

Uplink and downlink transmission based on the NR system may be based on the frame shown in FIG. 7. At this time, one wireless frame has a length of 10ms and can be defined as two 5ms half-frames (HF). One half-frame can be defined as five 1ms subframes (SF). One subframe is divided into one or more slots, and the number of slots in a subframe may depend on subcarrier spacing (SCS). At this time, each slot may include 12 or 14 OFDM(A) symbols depending on the cyclic prefix (CP). When normal CP (normal CP) is used, each slot may include 14 symbols. When extended CP (extended CP) is used, each slot may include 12 symbols. Here, the symbol may include an OFDM symbol (or CP-OFDM symbol) and an SC-FDMA symbol (or DFT-s-OFDM symbol).

Table 1 shows the number of symbols per slot according to SCS, the number of slots per frame, and the number of slots per subframe when a general CP is used, and Table 2 shows the number of symbols per slot according to SCS when an extended CSP is used. Indicates the number of symbols, the number of slots per frame, and the number of slots per subframe.

In Tables 1 and 2, Nslotsymb represents the number of symbols in a slot, Nframe,μslot represents the number of slots in a frame, and Nsubframe,μslot may represent the number of slots in a subframe.

Additionally, in a system to which this specification is applicable, OFDM(A) numerology (eg, SCS, CP length, etc.) may be set differently between a plurality of cells merged into one terminal. Accordingly, the (absolute time) interval of a time resource (e.g., SF, slot, or TTI) (for convenience, referred to as a time unit (TU)) composed of the same number of symbols may be set differently between merged cells.

NR can support multiple numerologies (or subcarrier spacing (SCS)) to support various 5G services. For example, if SCS is 15kHz, it supports wide area in traditional cellular bands, and if SCS is 30kHz/60kHz, it supports dense-urban, lower latency. And it supports a wider carrier bandwidth, and when the SCS is 60kHz or higher, it can support a bandwidth greater than 24.25GHz to overcome phase noise.

The NR frequency band is defined as two types (FR1, FR2) of frequency range. FR1 and FR2 can be configured as shown in the table below. Additionally, FR2 may mean millimeter wave (mmW).

Additionally, as an example, the above-described numerology may be set differently in a communication system to which this specification is applicable. As an example, a terahertz wave (THz) band may be used as a higher frequency band than the above-described FR2. In the THz band, the SCS can be set larger than the NR system, and the number of slots can also be set differently, and is not limited to the above-described embodiment.

Figure 8 is a diagram showing a slot structure applicable to this specification.

One slot includes multiple symbols in the time domain. For example, in the case of normal CP, one slot includes 7 symbols, but in the case of extended CP, one slot may include 6 symbols. A carrier includes a plurality of subcarriers in the frequency domain. RB (Resource Block) can be defined as multiple (eg, 12) consecutive subcarriers in the frequency domain.

Additionally, BWP (Bandwidth Part) is defined as a plurality of consecutive (P)RBs in the frequency domain and may correspond to one numerology (e.g., SCS, CP length, etc.).

A carrier wave may contain up to N (e.g., 5) BWPs. Data communication is performed through an activated BWP, and only one BWP can be activated for one terminal. Each element in the resource grid is referred to as a Resource Element (RE), and one complex symbol can be mapped.

6G communication system

6G (wireless communications) systems require (i) very high data rates per device, (ii) very large number of connected devices, (iii) global connectivity, (iv) very low latency, (v) battery- The goals are to reduce the energy consumption of battery-free IoT devices, (vi) ultra-reliable connectivity, and (vii) connected intelligence with machine learning capabilities. The vision of the 6G system can be four aspects such as “intelligent connectivity”, “deep connectivity”, “holographic connectivity”, and “ubiquitous connectivity”, and the 6G system can satisfy the requirements as shown in Table 4 below. In other words, Table 4 is a table showing the requirements of the 6G system.

At this time, the 6G system includes enhanced mobile broadband (eMBB), ultra-reliable low latency communications (URLLC), massive machine type communications (mMTC), AI integrated communication, and tactile communication. tactile internet, high throughput, high network capacity, high energy efficiency, low backhaul and access network congestion, and improved data security. It can have key factors such as enhanced data security.

Figure 9 is a diagram showing an example of a communication structure that can be provided in a 6G system applicable to this specification.

Referring to Figure 9, the 6G system is expected to have simultaneous wireless communication connectivity 50 times higher than that of the 5G wireless communication system. URLLC, a key feature of 5G, is expected to become an even more mainstream technology in 6G communications by providing end-to-end delays of less than 1ms. At this time, the 6G system will have much better volume spectrum efficiency, unlike the frequently used area spectrum efficiency. 6G systems can provide very long battery life and advanced battery technologies for energy harvesting, so mobile devices in 6G systems may not need to be separately charged. Additionally, new network characteristics in 6G may include:

- Satellites integrated network: 6G is expected to be integrated with satellites to serve the global mobile constellation. Integration of terrestrial, satellite and aerial networks into one wireless communications system could be critical for 6G.

- Connected intelligence: Unlike previous generations of wireless communication systems, 6G is revolutionary and will update the wireless evolution from “connected things” to “connected intelligence.” AI can be applied at each stage of the communication procedure (or each procedure of signal processing, which will be described later).

- Seamless integration wireless information and energy transfer: 6G wireless networks will deliver power to charge the batteries of devices such as smartphones and sensors. Therefore, wireless information and energy transfer (WIET) will be integrated.

- Ubiquitous super 3-dimemtion connectivity: Connectivity of drones and very low Earth orbit satellites to networks and core network functions will create super 3D connectivity in 6G ubiquitous.

In the above new network characteristics of 6G, some general requirements may be as follows.

- Small cell networks: The idea of small cell networks was introduced to improve received signal quality resulting in improved throughput, energy efficiency and spectral efficiency in cellular systems. As a result, small cell networks are an essential feature for 5G and Beyond 5G (5GB) communications systems. Therefore, the 6G communication system also adopts the characteristics of a small cell network.

- Ultra-dense heterogeneous network: Ultra-dense heterogeneous networks will be another important characteristic of the 6G communication system. Multi-tier networks comprised of heterogeneous networks improve overall QoS and reduce costs.

- High-capacity backhaul: Backhaul connections are characterized by high-capacity backhaul networks to support high-capacity traffic. High-speed fiber and free-space optics (FSO) systems may be possible solutions to this problem.

- Radar technology integrated with mobile technology: High-precision localization (or location-based services) through communication is one of the functions of the 6G wireless communication system. Therefore, radar systems will be integrated with 6G networks.

- Softwarization and virtualization: Softwarization and virtualization are two important features that are fundamental to the design process in 5GB networks to ensure flexibility, reconfigurability, and programmability. Additionally, billions of devices may be shared on a shared physical infrastructure.

This specification proposes a CSI feedback procedure and a corresponding device according to a multi-user-multi-stream CSI network structure to solve the problem of increasing signaling overhead due to multiple gradients.

Hereinafter, in this specification, “CSI network” may mean an artificial neural network that compresses and reconstructs CSI based on deep learning.

Below, to help understand the method proposed in this specification, the general details of a CSI network that compresses and reconstructs CSI based on deep learning will first be described.

Recently, various evolutions have been made in the architecture of CSI networks. Figure 10 is a diagram showing an example of a CSI network structure.

Recently, various evolutions have been made in the architecture of the CSI network.

Figure 10 shows an example of a neural network structure for CSI feedback. Figure 10 illustrates CsiNet, an example of a CSI network structure. Referring to FIG. 10, the CSI network can be viewed as consisting of a CSI encoder 1810 and a CSI decoder 1820. For example, in the case of downlink where data transmission is performed from the base station to the UE, the base station may operate as a transmitter and the UE may operate as a receiver. In the downlink, the CSI encoder may be operated by the UE, which is a receiver, and the CSI decoder may be operated by the base station, which is a transmitter. In this disclosure, the case of downlink communication is assumed for convenience of explanation, but various embodiments described later are not limited to downlink and can be applied to other links such as uplink and sidelink.

The CSI encoder included in the UE can compress information about channel conditions. Compressed information, which is the output of the CSI encoder, is transmitted to the base station through uplink feedback. The base station inputs the received compressed information to the CSI decoder, and the CSI decoder can restore information about the UE's channel state. In the present disclosure, for convenience of explanation, the compressed information that is the output of the CSI encoder and the input of the CSI decoder may be referred to as a CSI feedback signal, CSI feedback information, or other terms with equivalent technical meaning. . In the present disclosure, the CSI feedback signal may take the form of a bit stream. Here, a bit string refers to a sequence composed of binary digits or bits of 0 or 1, rather than a vector composed of floating point numbers.

Figures 11 and 12 are diagrams showing an example of a CSI network structure.

More specifically, FIGS. 11 and 12 show an example of CSI feedback bit streams accumulable in a decoder. It was difficult for the CSI network of FIG. 11 to support a variable feedback rate through the same neural network (NN) structure and model (parameter set) as in FIG. 12. In other words, in order to support multiple feedback rates in CSI feedback, models (parameter sets) for multiple encoders and decoders were needed.

On the other hand, in the case of the CSI feedback method shown in FIG. 12, multiple feedback streams can be transmitted from the encoder-side to the decoder-side. At this time, the number of feedback streams may vary depending on the feedback rate. At the decoder-end, CSI can be restored based on a signal that adds all or part of the feedback streams generated at the encoder-end. In this specification, “added” in expressions for CSI feedback signals such as “added before being input to the decoder NN,” “added and input to the decoder NN,” or “added to and input to the decoder NN.” The technique encompasses not only summation, but also weighted sum and weighted average. In this specification, for convenience of explanation, a CSI network in which multiple feedback streams can be transmitted from encoder-side to decoder-side is referred to as a “multi-stream CSI network.”

In the NN structure related to CSI, there is a multi-stream CSI network in which the UE can transmit a plurality of feedback signals to the BS. Although a multi-stream structure was needed to support variable feedback rates, an encoder structure of a CSI network that outputs multiple streams may exist for other purposes. The most common multi-stream CSI networks are considered herein.

Referring to FIG. 11, which relates to a general existing CSI network structure, an encoder neural network, a decoder neural network, and CSI feedback signals between the encoder neural network and the decoder neural network are shown. In Figure 11, three cases are shown according to the number of feedback bits. In each of the three cases, the vertical length of the CSI feedback bit stream can be expressed differently depending on the number of transmitted feedback bits, and the encoder neural network is used. The length of the right side and the left side of the green trapezoid that makes up the decoder neural network can also be expressed differently. In general, in the existing CSI network structure, the dimensions of the output of the encoder neural network and the input of the decoder neural network vary depending on the feedback rate, so the encoder Neural network and decoder The structure of the neural network itself may vary.

On the other hand, referring to FIG. 12, in the CSI feedback method of FIG. 12, different CSI feedback streams (feedback bit streams) may be added to each other before being input to the decoder neural network of the CSI network. Therefore, the input dimension of the decoder neural network can be maintained as is, the structure of the decoder neural network can be maintained as is, and furthermore, the model (parameter set) of the decoder neural network can also be maintained as is. It can be maintained. Figure 12 shows that the same decoder neural network model can always be used regardless of the number of CSI feedback streams added before being input to the decoder neural network. By adding it to the decoder neural network, the number of feedback bits increases in proportion to the number of input CSI feedback bit streams. For example, if the length of the CSI feedback bit stream that can be individually input to the decoder neural network is 256 bits, as the number of CSI feedback bit streams becomes 2, 3, and 4, the number of feedback bits is 512 and 768, respectively. , increases to 1024. Meanwhile, even if the same decoder neural network model is used, CSI reconstruction performance may improve as the number of CSI feedback bit streams added to the decoder neural network and input increases. In FIG. 12, as the number of CSI feedback bit streams increases, it is expressed that CSI reconstruction performance improves by expressing that the images (2391/2992/2393/2394) of FIG. 15 are restored with higher resolution. However, as shown in FIG. 12, the improvement in reconstruction performance expressed as an example of an image is only an example, and information such as the CSI matrix that can be restored in an actual CSI network is recognized by the human eye like the image in FIG. 12. It can be difficult to recognize. In addition, expressing the increase in CSI reconstruction performance as an increase in image resolution is only an example, and it may not be concluded that CSI reconstruction performance is improving in terms of image resolution. This is because, due to the nature of deep learning, it is difficult to explain exactly what the different CSI feedback streams input to the decoder neural network mean and what role they play.

Returning to Figure 12, CSI feedback streams playing different roles, which are added before being input to the decoder neural network, are expressed with different symbols. The CSI feedback bit stream 2301 may be a signal capable of CSI reconstruction even if it is independently input to the decoder neural network. Meanwhile, the CSI feedback bit stream 2302 must be added to the CSI feedback bit stream 2301 and input to the decoder neural network to enable CSI reconstruction. At this time, considering that the CSI feedback bit streams of different roles can be added and input to the decoder neural network as shown in FIG. 12, the CSI feedback bit streams of different roles are also referred to as CSI feedback bit streams of “different levels”. It can be understood. As shown in Figure 12, the structure of the encoder neural network, which plays a different role and outputs a CSI feedback bit stream that can be added to the decoder neural network and input, will be called ABC-Net ("Accumulable feature extraction Before skip Connection" Network). You can.

For purposes such as multiuser precoding, a situation in which CSI feedback signals from multiple UEs are transmitted to the BS can be considered. For convenience of explanation, in this specification, when multiple UEs exist, the NN in which the CSI feedback signal output from the encoder of each UE is input to the decoder of the BS is referred to as a “multiuser CSI network”. It is called.

For multi-user CSI networks, a scalable decoder architecture can be considered for multi-user precoding. Unlike the scalable decoder architecture, in the case of the decoder NN architecture for an end-to-end precoding system, feedback bits for a random user k cast

In this case, feedback bits collected from all K users

Receives all as input and creates a precoding matrix

is generated as output. This decoder NN structure is not at all scalable to K because the input and output size changes depending on the number of users K, and because it is not designed with special expert knowledge in mind, precoding performance cannot be further improved. There is room.

Figure 13 is a diagram showing an example of a decoder structure. Referring to FIG. 13, the decoder structure of FIG. 13 is such that the decoder NN 1330 receives as input all feedback bits collected from all K users in only one NN, and the entire precoding matrix (i.e., all precoding vectors) Instead of this structure being output at once, there is a separate decoder NN for each user, and a total of K NNs are structured in parallel. In the decoder structure of FIG. 13, the decoder NN for each user generates as output a precoding vector for only the user related to the decoder NN, and also requires a signal dedicated to that user as an input. A total of two signals are input to each decoder for each user, one of the two input signals is a feedback bit for the corresponding user, and the remaining input signal will be described later.

The second input of the decoder NN for each user (i.e., the kth user) is

(

) added together

It is a signal related to .

In the calculation of , the addition operation is an arithmetic operation, not a bitwise operation. Even if the CSI feedback signal is given as a bit stream, the addition operation may not be performed bitwise. At this time, an appropriate scaling factor (eg,

) can be multiplied. The k-th decoder NN is the k'-th user (

), the input

Considering

creates . In other words, the second signal input to the decoder NN can be interpreted as corresponding to inter-user interference.

In summary, in the case of Figure 13, as the input signal of the k-th decoder NN, the feedback bit of the k-th user

In addition to (1320), the sum of the feedback bits for all other users except the k-th user

The characteristic is that is used. Meanwhile, K parallel decoder NNs share the same structure and parameter set (weights and biases). In other words, decoders for different users use the exact same NN. Therefore, if there is only one decoder NN and the input signals for each user are sequentially input to the only decoder NN, it can be interpreted that the precoding vector for the corresponding user is sequentially output as an output. When training the parameter set of a decoder NN, this perspective may be effective. Since the decoder NN for each user is the same, all K decoder NNs are learned by sharing a common parameter set. That is, shared common parameters are learned.

Previously, representative examples of multi-user CSI networks and multi-stream CSI networks were described. In a multi-stream CSI network, multiple streams can be transmitted from the UE to the BS as CSI feedback, and in a multi-user CSI network, CSI feedback signals from multiple users can be transmitted to the BS. Therefore, a situation can be considered where CSI feedback signals are delivered to the BS from a plurality of UEs, but the CSI feedback from each UE is composed of multiple streams and delivered to the BS. In this specification, multiple feedback streams can be delivered to the BS from each UE, and the entire encoder and decoder NN in which CSI feedback is simultaneously provided to the BS from multiple UEs is used. It is defined as “multiuser CSI network with multiple feedback streams” or “multiuser multi-stream CSI network”.

In this specification, the decoder NN structure of a multiuser multi-stream CSI network is considered. Hereinafter, this specification assumes that the number of users is K, that S feedback streams can be delivered to the BS from each user, and that the ith feedback stream of the kth user is

It is expressed as. At this time, each feedback stream

(

about

is equivalent to the bit stream of B bits.

This is,

It can be. That is, each bidback stream

against vector

Assume that can be expressed as B components. In other words, each feedback stream

The dimension of can be seen as B. The decoder NN in the BS-stage typically has

can be seen as being input.

Additionally, in this specification, it is assumed that at least in the first layer of the decoder NN, operations are performed only between components at the same location for different feedback streams. In other words, it is assumed that operations between components existing at different locations in different feedback streams can be performed starting from the second layer of the decoder NN.

The most general case of a multiuser multi-stream CSI network can be expressed as shown in FIG. 14. Hereinafter, in this specification, the part of the first layer of the decoder NN excluding the nonlinear activation function is referred to as “decoder head”. In other words, the decoder head refers to the part of the decoder NN corresponding to the range immediately before the activation function for the first layer is applied. In addition, hereinafter, in this specification, the decoder NN excluding the decoder head may be referred to as the decoder body or the main decoder NN, and the nonlinear activation function excluded from the decoder head may be included in the main decoder NN, so the decoder body is It can start.

Referring to FIG. 14, the part represented by the black box (1420) corresponds to the decoder head, and the remaining decoder NN part excluding the decoder head corresponds to the decoder body (1430). In this specification, it is assumed that a total of J input vectors are input to the main decoder NN. Additionally, unless otherwise specified, the decoder body is expressed as (main) decoder NN. The jth input vector to Decoder NN is

It is expressed as, and depending on the assumption,

It can also be expressed as B components.

The dimension of can be interpreted as B,

It can be.

When following the assumptions in this specification described above, any decoder NN of a multiuser multi-stream CSI network can be expressed as divided into a decoder head and a main decoder NN as shown in FIG. 14. At this time, the decoder head input vector to the main decoder NN.

go

It can be calculated as follows. here,

is a learnable parameter that can be obtained or updated through learning of NN.

In wireless communication, since a mobile situation is generally assumed where the UE is moving, the distribution (statistics) of the wireless channel may change frequently. At this time, the encoder and decoder NN models of the CSI network must be changed according to changes in channel distribution (statistics), but it may be realistically difficult for the UE to download the changed model (parameter set) each time, and also, the UE must change the channel distribution (statistics) ), it is difficult for the UE to make various parameter sets readily available by learning in advance a model (parameter set) for all possible cases. Therefore, online learning can be considered, and this specification also considers online learning situations. In order for a multi-stream CSI network in which multiple feedback streams can be transmitted from the encoder-side (UE) to the decoder-side (BS) to be learned through online learning, Multiple gradient vectors corresponding to the number of feedback streams must be transferred from decoder-side (BS) to encoder-side (UE). This is because backpropagation requires a gradient vector for each feedback stream. Therefore, a multi-stream CSI network has a problem in that signaling overhead for online learning increases by the number of feedback streams compared to a CSI network that supports only a single feedback bit stream. 15 and 16 show backpropagation when the number S of feedback streams from each UE is 2 in a multi-stream CSI network, compared to the case where only a single feedback stream is supported from each UE (S = 1). This shows the problem of doubling the signaling overhead for . Generally, for online learning of multi-stream CSI networks, the problem of multiple gradients for multiple feedback streams arises. Therefore, a multi-user multi-stream CSI network structure that can solve the problem of increased signaling overhead due to multiple gradients is needed.

Below, we describe a multi-user multi-stream CSI network structure that can solve the problem of increased signaling overhead due to multiple gradients as the number of feedback streams increases.

First, we will describe how to configure an encoder NN that outputs a bit stream as a CSI feedback signal. In general, the output of a fully-connected (FC) layer can be viewed as a vector made up of real numbers.

The output of the FC layer is a bipolar vector equivalent to a bit stream of B bits.

sign function as the activation function of the FC layer so that can be output

There may be a way to apply . Sign function is also called signum function and is defined as the equation below.

Various methods can be applied to ensure that the CSI feedback signal is output from the encoder NN in the form of a bit stream and input to the decoder NN. however,

The slope (differentiation coefficient) of the function is 0 in most of the domain, and differentiation is not possible in the remaining parts. Therefore, the gradient disappears in almost all areas, making backpropagation difficult, which makes learning difficult. As a surrogate for backpropagation to solve difficulties in learning

A Straight-Through Estimator (STE) can be used to replace the function. In the forward pass, the original quantized activation function is passed, and STE is used only in backpropagation. At this time,

A function that appropriately approximates the function, but is differentiable in the required region and whose differential coefficient is no longer 0, making the gradient non-trivial can be used as an STE. For example,

The function can be approximated and replaced as shown in the equation below.

here,

is the annealing factor in the l-th epoch and is the slope of the sigmoid function that gradually increases as training progresses. As shown in the above equation, a STE that appropriately approximates the original function through the sigmoid function is called a sigmoid-adjusted STE. As the slope of the sigmoid function increases, it better approximates the signum function. In addition, in the above equation, a method is used in which the slope of the sigmoid function increases as the number of epochs increases and learning progresses. This method is called the slope-annealing trick, and the performance of STE is further improved by using the slop-annealing method. It can be improved.

Hereinafter, in this specification, for convenience of explanation, it is basically assumed that “sigmoid-adjusted STE with slope annealing” is applied. However, the present invention is not limited to the sigmoid-adjusted STE of the slope annealing method.

Additionally, in this specification, in order to distinguish it from the gradient in the general case where the surrogate of backpropagation is not replaced by STE, the gradient obtained by passing STE in backpropagation is called coarse gradient. At this time, in order to operate with a learning method such as gradient descent, the coarse gradient obtained by the STE-modified chain rule can be transmitted from the BS to the UE.

Gradient in this specification may have a meaning encompassing coarse gradient. In other words, there is no distinction between general gradient and coarse gradient. However, since this specification basically assumes that “sigmoid-adjusted STE with slope annealing” is applied for convenience of explanation, the gradient transmitted from the BS to the UE in this specification can be understood as a coarse gradient. However, the methods proposed in this specification are not limited to coarse gradients, and the methods proposed in this specification can be applied even in situations where there is a general gradient to which STE is not applied.

Two-layered decoder head where common gradient can be back-propagated

Below, the multi-user multi-stream CSI network structure will be described.

Figure 17 is a diagram showing an example of a multi-user multi-stream CSI network structure proposed in this specification. The multi-user multi-stream CSI network structure described in FIG. 14 can be modified as shown in FIG. 17. At this time, the encoder 1710 and decoder NN 1730, excluding the decoder head 1720, can be maintained without being modified compared to the multi-user multi-stream CSI network structure described in FIG. 14. That is, only the decoder head represented by the black box 1420 in FIG. 14 can be modified as represented by 1720/1721 in FIG. 17. In other words, this specification proposes a decoder head structure for a multi-user multi-stream CSI network. Any decoder head (1420) as shown in Figure 14

After receiving input,

outputs. Therefore, given an arbitrary decoder head, the decoder head is set to a parameter set

It can be expressed as Accordingly, any given parameter set for the decoder head

This can be considered. arbitrary parameter set

Given this, all

For matrix

can be defined. At this time, the matrix

A low-rank approximation of can be found, for example, a singular value decomposition (SVD) method can be used to find a low-rank approximation. given a random matrix

about

satisfying

and

can be found. Generally

ramen, procession

While the rank of is at most S, the matrix

The rank of may be 1. In other words, you can find an approximation of rank 1. That is, all

For matrix

can be approximately expressed

and

can be found. Therefore, given an arbitrary parameter set

via approximation to

It can be expressed as follows.

Therefore, any decoder head as shown in Figure 14

It can be approximately expressed as, and can be expressed as decoder head (1720/1710) in FIG. 17.

Before explaining the characteristics of the decoder head structure proposed in this specification in more detail, the concept of "vector neuron" may be defined in this specification, and the concept of artificial neural network (NN) may be used to define a vector neuron. First, we will explain the concept of a neuron, which can be considered a unit structure.

Generally, in a NN, a neuron has N+1 scalar input variables.

Given this, each input

weight, which is a learnable parameter in

This multiplied and added scalar value

activation function in

This refers to the structure in which the applied value is output. In other words, the input of the neuron is a plurality of scalar values.

And the output of the neuron is also a scalar value.

am. At this time, the variable

When fixed to , it is a learnable parameter.

can be called bias. However, in this specification, for convenience of explanation, bias is not separately considered.

Meanwhile, N vector input variables with the same dimension

and the corresponding scalar learnable parameters

For , a vector is a linear combination of input vectors.

Component-wise activation function

The output vector to which this is applied

The NN structure that outputs can be defined as a vector neuron. In other words, the vector neuron defined in this specification can be understood as extending the concept of a general neuron from a scalar perspective to a vector perspective, and when the dimensions of the input vector and output vector are 1, a vector neuron ( A vector neuron may be the same concept as a general neuron.

The decoder head represented by a black box (1420) in FIG. 14 can be interpreted as a NN with one layer consisting of a total of J vector neurons. Each vector neuron in the decoder head is

Receives 1 vector as input and produces 1 vector

outputs. At this time, the activation function of each vector neuron present in the decoder head is omitted, so the activation function can be applied in the main decoder NN instead of being applied in the decoder head.

Meanwhile, the characteristics of the decoder head structure proposed in this specification can be summarized as follows.

(1) The proposed decoder head can be interpreted like a NN structure consisting of two layers.

(2) It can be seen that there are as many vector neurons as the number of users K in the first layer of the two layers.

(3) Each vector neuron in the first layer receives only feedback streams from a specific user.

(4) It can be seen that the second layer of the two layers contains as many vector neurons as the number J of input vectors for the (main) decoder NN.

(5) The output of each vector neuron of the second layer is input to the decoder NN.

Unlike the decoder head represented by the black box (1420) in FIG. 14 composed of one layer, the decoder head (1720/1721) shown in FIG. 17 is composed of two layers. In the first layer, there are vector neurons equal to the number of users K, and in the second layer, there are vector neurons equal to the number J of input vectors to the decoder NN.

Each of the K vector neurons 1720 present in the first layer of the decoder head of FIG. 17 may correspond to a specific user. The vector neuron corresponding to the kth user has S feedback streams for the kth user.

Receives 1 vector as input

can be output. in other words,

is a linear combination of the feedback streams for the kth user. Although not shown in Figure 17,

A vector with an appropriate activation function applied to

may be output.

Each of the J vector neurons 1721 present in the second layer of the decoder head in FIG. 17 inputs one specific input vector among the J input vectors to the (main) decoder NN 1730. vector). The vector neuron corresponding to the j-th input vector to the Decoder NN (1730) is the K vectors output from the first layer.

takes as input and produces 1 vector

can be output. in other words,

is a linear combination of the output vectors output from the first layer. At this time,

An appropriate activation function may be applied, but depending on the meaning of decoder head defined in this specification, application of the activation function may be performed at the first part of the (main) decoder NN instead of the decoder head.

Additionally, as shown in Figure 17, when there is no activation function between the first and second layers of the decoder head, the two layers may be interpreted as one layer. The reason is that,

Therefore, the proposed decoder head structure is,

Receives 1 vectors and creates 1 vector

This is because it can be viewed as a single-layer NN consisting of J vector neurons that output . Previously, we mainly explained the case where no activation function exists between the first layer and the second layer of the decoder head, but an appropriate activation function may be applied between the first layer and the second layer, and in this case, the two layers of the decoder head are interpreted as one layer. Can not. The content proposed in this specification is not limited to the presence or absence of an activation function.

There can be two ways to obtain the model (parameter set) of the proposed decoder head. First, the first method is the model of the decoder head, which is already given

A new parameter set obtained by applying low-rank approximation to

This method replaces the existing decoder head and performs fine tuning for the entire multi-user multi-stream CSI network. At this time, in the fine tuning process, learning may proceed while the parameters for all or part of the NN except the decoder head are fixed. The second method may be to configure the decoder head with a proposed structure from the beginning and train it.

Signaling for online learning of proposed architecture

Below, a method for reducing signaling overhead for online learning using the decoder head structure proposed in this specification will be described.

Figure 18 is a diagram showing an example of an online learning method to which the decoder head structure proposed in this specification is applied.

Referring to Figure 18, when a two-layered decoder head is used, in the backpropagation procedure, common gradient information is used instead of multiple gradient vectors on the decoder-side ( BS) can be transmitted from encoder-side (UE). At this time, weights used for linear combination of multiple feedback streams for each user (kth user) with common gradient information.

It can also be transmitted from decoder-side (BS) to encoder-side (UE).

For example, in the forward pass procedure, multiple feedback streams

When transmitted from this kth user to the BS, the BS

can be calculated in the first layer of the proposed decoder head. After the learning process is performed, the gradient for each layer of the loss function L is transferred from the last layer of the decoder NN to the input of the decoder NN through backpropagation and can be calculated. thus,

The gradient of the loss function L for

can be calculated. Backpropagation continues to calculate the gradient for each layer of the encoder NN, and in order to be transmitted from the last layer of the encoder NN to the input of the decoder NN, the kth user must calculate the gradient vector for the S feedback streams. vectors)

You must know everything. At this time, S gradient vectors

Instead of everything being transmitted from the BS to the kth user, as shown in Figure 18

The gradient vector of the loss function L for

If only the signal is transmitted, the effect of reducing signaling overhead can be obtained. Therefore, BS is common gradient information.

with

can be transmitted to the kth user.

The kth UE received

and

Based on , gradient vectors for S feedback streams.

Since it can be calculated as follows, backpropagation can be performed and the gradient for each layer of the encoder NN can be calculated and transmitted. That is, a multi-user multi-stream CSI network can be learned. The signaling procedure shown in FIG. 18 can be repeated in batch units of online learning.

that must be delivered to the kth user from decoder-side (BS)

Can be quantized and transmitted in a pre-arranged manner between the kth UE and the BS, as in existing general digital communication. At this time, only relative/differential values based on weights in a specific order can be transmitted. For example,

based on this

is a relative ratio rather than the absolute value of

Only can be transmitted. also,

If the constraints of are satisfied, only weights other than weights in a specific order can be transmitted. For example,

weights excluding

Only can be transmitted. At this time, the kth UE receives

and constraint

The weight that was not transmitted based on

go

It can be calculated as follows. in other words,

Part or all of may be transmitted from the BS to the kth UE implicitly without being directly expressed. Common gradient information can also be transmitted in a pre-arranged manner between the UE and BS, including quantization, similar to signaling in existing general digital communication.

effect

The following effects can be achieved through the methods proposed in this specification.

First, signaling overhead for online learning of a multi-user multi-stream CSI network can be reduced.

Additionally, in the backpropagation procedure, the amount of information to be transmitted from the BS to the UE can be reduced by about 1/S. Here, S represents the number of feedback streams.

Figures 19 and 20 are diagrams showing the effect of the decoder structure proposed in this specification.

More specifically, referring to FIG. 19 for a case in which the decoder structure proposed in this specification is not applied, a general case in which the number of feedback streams is S and each bit stream is composed of B bits ( B>>S), for online learning of a multi-user multi-stream CSI network, the number of real numbers to be transmitted from the BS to each UE in the backpropagation procedure is as much as SXB. On the other hand, referring to FIG. 20 for the case of using the decoder head structure proposed in this specification, real numbers to be transmitted from the BS to each UE in the backpropagation procedure for online learning of a multi-user multi-stream CSI network. The number can be reduced to S+B. Here, dividing the number of real numbers to be transmitted from the BS to each UE by B, if the decoder head structure proposed in this specification is not applied, the (downlink) signaling overload required in a multi-user multi-stream CSI network is The head is S, but when the decoder head structure proposed in this specification is applied, the (downlink) signaling overhead required in a multi-user multi-stream CSI network is 1+.

, it can be seen that when the decoder head structure proposed in this specification is applied, the signaling overhead is reduced at a rate of about 1/S.

Figure 21 is a flowchart showing an example of how the channel state reporting method proposed in this specification is performed in a terminal.

First, the terminal receives configuration information related to CSI reporting from the base station (s2110).

Here, the setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI. do.

Next, the terminal receives a channel state information reference signal (CSI-RS) from the base station (S2120).

Afterwards, the terminal transmits the CSI to the base station (S2130).

Here, the CSI is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station.

Additionally, the terminal includes a transmitter for transmitting a wireless signal; A receiver for receiving wireless signals; at least one processor; and at least one computer memory operably connectable to the at least one processor and storing instructions that, when executed by the at least one processor, perform operations. At this time, the operations include the steps described in FIG. 21 above.

Additionally, the operations described in FIG. 21 may be stored in a non-transitory computer readable medium (CRM) that stores one or more instructions. The non-transitory computer-readable medium stores one or more instructions executable by one or more processors, and the one or more instructions cause the transmitting end to perform the operation described in FIG. 21.

Additionally, a device including one or more memories and one or more processors functionally connected to the one or more memories, wherein the one or more processors control the device to perform the operations described in FIG. 25 .

Figure 22 is a flowchart showing an example of how the channel state reporting method proposed in this specification is performed at the base station.

First, the base station transmits configuration information related to CSI reporting to each of a plurality of terminals (s2210).

Here, the setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI. do.

Next, the base station transmits a channel state information reference signal (CSI-RS) to each of the plurality of terminals (S2220).

Thereafter, the base station transmits the CSI to each of the plurality of terminals (S2230).

Here, the CSI is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station.

In addition, the base station includes a transmitter for transmitting wireless signals; A receiver for receiving wireless signals; at least one processor; and at least one computer memory operably connectable to the at least one processor and storing instructions that, when executed by the at least one processor, perform operations. At this time, the operations include the steps described in FIG. 22 above.

Additionally, the operations described in FIG. 22 may be stored in a non-transitory computer readable medium (CRM) that stores one or more instructions. The non-transitory computer-readable medium stores one or more instructions executable by one or more processors, and the one or more instructions cause the receiving end to perform the operation described in FIG. 22.

Additionally, a device including one or more memories and one or more processors functionally connected to the one or more memories, wherein the one or more processors control the device to perform the operations described in FIG. 22 .

The embodiments described above combine the components and features of the present invention in a predetermined form. Each component or feature should be considered optional unless explicitly stated otherwise. Each component or feature may be implemented in a form that is not combined with other components or features. Additionally, it is possible to configure an embodiment of the present invention by combining some components and/or features. The order of operations described in embodiments of the present invention may be changed. Some features or features of one embodiment may be included in other embodiments or may be replaced with corresponding features or features of other embodiments. It is obvious that claims that do not have an explicit reference relationship in the patent claims can be combined to form an embodiment or included as a new claim through amendment after filing.

Embodiments according to the present invention may be implemented by various means, for example, hardware, firmware, software, or a combination thereof. In the case of implementation by hardware, an embodiment of the present invention includes one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), and FPGAs ( It can be implemented by field programmable gate arrays, processors, controllers, microcontrollers, microprocessors, etc.

In the case of implementation by firmware or software, an embodiment of the present invention may be implemented in the form of a module, procedure, function, etc. that performs the functions or operations described above. Software code can be stored in memory and run by a processor. The memory is located inside or outside the processor and can exchange data with the processor through various known means.

It is obvious to those skilled in the art that the present invention can be embodied in other specific forms without departing from the essential features of the present invention. Accordingly, the above detailed description should not be construed as restrictive in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

The present invention has been described focusing on examples of application to 3GPP LTE/LTE-A and 5G systems, but it can be applied to various wireless communication systems in addition to 3GPP LTE/LTE-A and 5G systems.

Claims (15)

1. In a method for a plurality of terminals to report channel state information (CSI) in a wireless system, the method performed by one of the plurality of terminals includes,

   Receiving configuration information related to CSI reporting from the base station,

   The setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI;

   Receiving a channel state information reference signal (CSI-RS) from the base station;

   Including transmitting the CSI to the base station,

   The CSI is a method characterized in that it is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station.
2. According to claim 1,

   A method characterized in that the CSI input to the one element is output as an output value generated based on a linear combination of at least one information stream related to the CSI.
3. According to claim 2,

   The output value generated based on a linear combination of at least one information stream related to the CSI is used as an input for all or part of the elements constituting the second layer of the decoder neural network included in the base station. method.
4. According to claim 3,

   The elements constituting the first layer of the decoder neural network and the elements constituting the second layer of the decoder neural network are vector neurons that receive input vector variables and perform a linear combination operation on the input vector variables. A method characterized in that.
5. According to claim 4,

   A method characterized in that no activation function is applied between the first layer of the decoder neural network and the second layer of the decoder neural network.
6. According to claim 1,

   The method of claim 1, wherein each of the at least one weights is each associated with one of the at least one information stream.
7. According to claim 6,

   The method characterized in that the weight information is information about the relative ratios of the remaining weights to a specific weight among the at least one weight.
8. According to claim 6,

   A method characterized in that the weight information is information about remaining weights excluding information about a specific weight among the at least one weight.
9. According to claim 8,

   A method wherein information about a certain weight is determined based on the weight information under predefined constraints.
10. According to claim 1,

    The common gradient information is information about the gradient vector of a loss function for the output of a specific element related to a specific terminal among the plurality of terminals, among the elements constituting the first layer of the decoder neural network. A method characterized by:
11. In a wireless communication system, one terminal reports channel status together with multiple terminals,

    A transmitter for transmitting wireless signals;

    A receiver for receiving wireless signals;

    at least one processor; and

    at least one computer memory operably connectable to the at least one processor and storing instructions that, when executed by the at least one processor, perform operations;

    The above operations are:

    Receiving configuration information related to CSI reporting from the base station,

    The setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI;

    Receiving a channel state information reference signal (CSI-RS) from the base station;

    Including transmitting the CSI to the base station,

    The CSI is a terminal characterized in that it is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station.
12. A method for a base station to receive channel state information from a plurality of terminals in a wireless communication system includes:

    Transmitting setting information related to CSI reporting to each of the plurality of terminals,

    The setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI;

    Transmitting a channel state information reference signal (CSI-RS) to each of the plurality of terminals;

    Including receiving the CSI from each of the plurality of terminals,

    The CSI is a method characterized in that it is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station.
13. In a wireless communication system, a base station that receives channel state information from a plurality of terminals,

    A transmitter for transmitting wireless signals;

    A receiver for receiving wireless signals;

    at least one processor; and

    at least one computer memory operably connectable to the at least one processor and storing instructions that, when executed by the at least one processor, perform operations;

    The above operations are:

    Transmitting setting information related to CSI reporting to each of the plurality of terminals,

    The setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI;

    Transmitting a channel state information reference signal (CSI-RS) to each of the plurality of terminals;

    Including receiving the CSI from each of the plurality of terminals,

    The CSI is a base station, characterized in that it is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station.
14. In a non-transitory computer readable medium (CRM) storing one or more instructions,

    One or more instructions executable by one or more processors allow the terminal to:

    From the base station, receive configuration information related to CSI reporting,

    The setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI;

    Receive a channel state information reference signal (CSI-RS) from the base station;

    To the base station, transmit the CSI,

    The CSI is a non-transitory computer, characterized in that it is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station. Readable media.
15. A device comprising one or more memories and one or more processors functionally connected to the one or more memories,

    The one or more processors allow the device to:

    From the base station, receive configuration information related to CSI reporting,

    The setting information includes (i) common gradient information and/or (ii) weight information for at least one weight for linear combination of at least one information stream related to the CSI;

    Receive a channel state information reference signal (CSI-RS) from the base station;

    To the base station, transmit the CSI,

    The CSI is a device characterized in that it is used as an input for one element related to the one terminal among the elements constituting the first layer of the decoder neural network included in the base station.

Images