Method and apparatus for channel estimation in wireless communication system

Abstract

A method performed by a base station in a wireless communication system, according to embodiments of the present disclosure, comprises the steps of: transmitting configuration information about a sounding reference signal (SRS) to a terminal; receiving a first SRS from the terminal; and on the basis of the received first SRS, estimating a channel for wireless communication by using at least one artificial intelligence model.

Description

Method and apparatus for channel estimation in a wireless communication system

[0001] The present disclosure relates to a method and apparatus for channel estimation in a wireless communication system. Specifically, it relates to a method and apparatus for performing enhanced channel estimation using various auxiliary information.

[0002] Looking back at the evolution of wireless communication through successive generations, technologies have been developed primarily for human-oriented services, such as voice, multimedia, and data. Following the commercialization of 5G (5th Generation) communication systems, connected devices, which have been increasing explosively, are expected to be connected to communication networks. Examples of networked objects include vehicles, robots, drones, home appliances, displays, smart sensors installed in various infrastructures, construction machinery, and factory equipment. Mobile devices are expected to evolve into various form factors, such as augmented reality glasses, virtual reality headsets, and holographic devices. In the 6G (6th Generation) era, efforts are underway to develop improved 6G communication systems to connect hundreds of billions of devices and objects to provide diverse services. For this reason, 6G communication systems are being referred to as "beyond 5G" systems.

[0003] In the 6G communication system predicted to be realized around 2030, the maximum transmission speed is tera (i.e., 1,000 gigabit) bps (bit per second), and the wireless latency is 100 microseconds (μsec). In other words, compared to the 5G communication system, the transmission speed in the 6G communication system is 50 times faster, and the wireless latency is reduced to one-tenth.

[0004] To achieve such high data transmission speeds and ultra-low latency, 6G communication systems are being considered for implementation in the terahertz (THz) band (e.g., the 95 gigahertz (GHz) to 3 terahertz (3THz) band). Due to more severe path loss and atmospheric absorption phenomena compared to the millimeter wave (mmWave) band introduced in 5G, the importance of technologies capable of guaranteeing signal reach, or coverage, is expected to increase in the terahertz band. As key technologies to ensure coverage, new waveforms, beamforming, and multi-antenna transmission technologies such as massive Multiple-Input and Multiple-Output (MIMO), Full Dimensional MIMO (FD-MIMO), array antennas, and large-scale antennas, which are superior in terms of coverage compared to RF (Radio Frequency) devices, antennas, and OFDM (Orthogonal Frequency Division Multiplexing), must be developed. In addition, new technologies such as metamaterial-based lenses and antennas, high-dimensional spatial multiplexing technology using Orbital Angular Momentum (OAM), and Reconfigurable Intelligent Surface (RIS) are being discussed to improve the coverage of terahertz band signals.

[0005] In addition, to improve frequency efficiency and system network, development is underway in 6G communication systems for full duplex technology, in which uplink and downlink simultaneously utilize the same frequency resources at the same time; network technology that integrates satellites and HAPS (High-Altitude Platform Stations); network structure innovation technology that supports mobile base stations and enables network operation optimization and automation; dynamic spectrum sharing technology through collision avoidance based on spectrum usage prediction; AI-based communication technology that utilizes AI (Artificial Intelligence) from the design stage and internalizes end-to-end AI support functions to realize system optimization; and next-generation distributed computing technology that realizes services of complexity exceeding the limits of terminal computing capabilities by utilizing ultra-high performance communication and computing resources (Mobile Edge Computing (MEC), cloud, etc.). In addition, attempts are continuing to further strengthen connectivity between devices, further optimize networks, promote the softwareization of network entities, and increase the openness of wireless communication through the design of new protocols to be used in 6G communication systems, the implementation of hardware-based security environments, the development of mechanisms for the safe utilization of data, and the development of technologies regarding privacy maintenance methods.

[0006] Due to the research and development of such 6G communication systems, it is expected that a new dimension of hyper-connected experience will become possible through the hyper-connectivity of 6G communication systems, which encompasses not only connections between objects but also connections between people and objects. Specifically, it is projected that 6G communication systems will enable the provision of services such as truly immersive eXtended Reality (XR), high-fidelity mobile holograms, and digital replicas. Furthermore, services such as remote surgery, industrial automation, and emergency response, which are provided through 6G communication systems with enhanced security and reliability, will be applied in various fields including industry, healthcare, automotive, and home appliances.

[0007] The present disclosure may have one objective of providing a method and apparatus with improved channel estimation performance in a wireless communication system.

[0008] A method performed by a base station in a wireless communication system according to embodiments of the present disclosure comprises the steps of transmitting configuration information for a sounding reference signal (SRS) to a terminal, receiving a first SRS from the terminal, and estimating a channel for wireless communication using at least one artificial intelligence model based on the received first SRS.

[0009] In one embodiment, the step of estimating a channel for wireless communication includes the step of generating a first channel estimation matrix using a first artificial intelligence model based on a first SRS, and the step of inputting a second channel and a first channel estimated based on a second SRS received prior to the first SRS into a second artificial intelligence model to output a third channel, and the first artificial intelligence model may include a model for removing noise from the input.

[0010] In one embodiment, the step of estimating a channel for wireless communication may include the step of generating a first feature vector using a third artificial intelligence model based on a first SRS, and the step of generating a fourth channel estimation matrix by inputting the first feature vector and a second feature vector generated based on a second SRS received prior to the first SRS into a fourth artificial intelligence model.

[0011] In one embodiment, the method further comprises the step of obtaining auxiliary information for estimating a channel, and the auxiliary information may include measurement information measured at a terminal or base station, information generated based on the measurement information, or an indicator regarding a method for estimating a channel. The measurement information may include at least one of RSRP (reference signal received power), RSRQ (reference signal received quality), CQI (channel quality indicator), the speed of the terminal, or Doppler shift. Additionally, the indicator regarding a method for estimating a channel may include a first indicator for indicating whether to estimate a channel based on a second SRS received prior to a first SRS. Furthermore, the first indicator may be determined based on a value output by an outlier detection model learned based on a normal SRS, which receives the second SRS as input. Additionally, the first indicator may be determined based on a comparison of a measured Doppler shift with an arbitrary threshold.

[0012] In one embodiment, an indicator for a method of estimating a channel includes a second indicator for indicating any one of a plurality of artificial intelligence models stored in a base station, and the second indicator can be determined based on measurement information.

[0013] In one embodiment, the artificial intelligence model may include an encoder-decoder or some of its components intended for neural network-based noise removal. For example, the artificial intelligence model may include a convolutional neural network-based autoencoder for noise removal.

[0014] The method and apparatus according to the embodiments of the present disclosure can improve channel estimation performance in a wireless communication system.

[0015] Specifically, the embodiments of the present disclosure can derive improved channel estimation performance by considering the SRS transmitted from past slots together.

[0016] Specifically, the embodiments of the present disclosure can derive improved channel estimation performance by considering an artificial intelligence model for channel estimation.

[0017] Specifically, the embodiments of the present disclosure can derive improved channel estimation performance by considering various auxiliary information for channel estimation.

[0018] The effects obtainable from the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art to which the present disclosure belongs from the description below.

[0019] The features and advantages of the embodiments of the present disclosure will become more apparent from the following description together with the accompanying drawings.

[0020] FIG. 1 illustrates a wireless communication system according to embodiments of the present disclosure.

[0021] FIG. 2 is a drawing for explaining the structure of a terminal according to embodiments of the present disclosure.

[0022] FIG. 3 is a drawing for explaining the structure of a network entity (or base station) according to embodiments of the present disclosure.

[0023] FIG. 4 illustrates an example of SRS transmission of a terminal and channel estimation of a base station according to embodiments of the present disclosure.

[0024] FIG. 5 illustrates an example of SRS reception and data generated in the antenna / frequency domain according to embodiments of the present disclosure.

[0025] FIG. 6 illustrates a method for estimating a channel using an artificial intelligence model according to embodiments of the present disclosure.

[0026] FIGS. 7A and 7B illustrate a method for estimating a channel using an artificial intelligence model according to embodiments of the present disclosure.

[0027] FIG. 8 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0028] FIG. 9 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0029] FIG. 10 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0030] FIG. 11 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0031] FIGS. 12A and FIGS. 12B illustrate a method for estimating a channel according to embodiments of the present disclosure.

[0032] FIGS. 13A and FIGS. 13B illustrate a method for estimating a channel considering a Doppler shift according to embodiments of the present disclosure.

[0033] FIG. 14 illustrates an example of the operation of an artificial intelligence model for estimating a channel according to embodiments of the present disclosure.

[0034] FIG. 15 illustrates an example of the operation of an artificial intelligence model for estimating a channel according to embodiments of the present disclosure.

[0035] FIG. 16 illustrates an example of the operation of a memory for estimating a channel according to embodiments of the present disclosure.

[0036] FIG. 17 illustrates an example of the operation of a base station for estimating a channel according to embodiments of the present disclosure.

[0037] FIG. 18 illustrates an example of the operation of a terminal for estimating a channel according to embodiments of the present disclosure.

[0038] Embodiments of the present disclosure may solve the problems and / or disadvantages described above and provide the advantages described below. One aspect of the present disclosure may provide a network entity (or node) and a method of communication thereof in a wireless communication system.

[0039] The terms used in this disclosure are used merely to describe specific embodiments and are not intended to limit the scope of other embodiments. A singular expression may include a plural expression unless the context clearly indicates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as generally understood by those skilled in the art described in this disclosure. Terms used in this disclosure that are defined in a general dictionary may be interpreted as having the same or similar meaning as they have in the context of the relevant technology, and are not to be interpreted in an ideal or overly formal sense unless explicitly defined in this disclosure. In some cases, even terms defined in this disclosure are not to be interpreted to exclude the embodiments of this disclosure.

[0040] The various embodiments of the present disclosure described below illustrate a hardware-based approach. However, since the various embodiments of the present disclosure include techniques using both hardware and software, the various embodiments of the present disclosure do not exclude a software-based approach.

[0041] Additionally, various embodiments of the present disclosure describe various embodiments using terms used in some communication standards (e.g., 3GPP (3rd generation partnership project)), but this is merely for illustrative purposes. Various embodiments of the present disclosure can be easily modified and applied to other communication systems.

[0042] Various embodiments of the present disclosure are described below.

[0043] FIG. 1 illustrates a wireless communication system according to embodiments of the present disclosure.

[0044] FIG. 1 illustrates a base station (110), a first terminal (120), and / or a second terminal (130) as part of nodes utilizing a wireless channel in a wireless communication system. FIG. 1 illustrates only one base station, but this is merely an example. The wireless communication system of FIG. 1 may include other base stations identical or similar to the base station (110).

[0045] A base station (110) is a network infrastructure that provides wireless access to terminals (120, 130). The base station (110) has coverage defined as a certain geographical area based on the distance at which it can transmit signals. In addition to being a base station, the base station (110) may be referred to as an 'access point (AP)', 'evolved Node B (eNB)', 'next generation node B (gNB)', '5G node (5th generation node)', 'wireless point', 'transmission / reception point (TRP)', or other terms having an equivalent technical meaning.

[0046] Each of the first terminal (120) and the second terminal (130) is a device used by a user and can perform communication with the base station (110) via a wireless channel. At least one of the first terminal (120) or the second terminal (130) can be operated without user involvement. For example, at least one of the first terminal (120) or the second terminal (130) may be a device that performs machine type communication (MTC) and may not be carried by the user. Each of the first terminal (120) and the second terminal (130) may be referred to as 'user equipment (UE)', 'mobile station', 'subscriber station', 'customer premises equipment (CPE)', 'remote terminal', 'wireless terminal', 'electronic device', or 'user device' or other terms having an equivalent technical meaning.

[0047] The base station (110), the first terminal (120), and the second terminal (130) can transmit and / or receive wireless signals in a millimeter wave (mmWave) band (e.g., 28 GHz, 30 GHz, 38 GHz, 60 GHz). At this time, to improve channel gain, the base station (110), the first terminal (120), and / or the second terminal (130) can perform beamforming.

[0048] Beamforming may include transmitting beamforming and / or receiving beamforming. That is, the base station (110), the first terminal (120), and / or the second terminal (130) may give directivity to the transmitted signal or the received signal. To give directivity to the received signal, the base station (110) and / or the terminals (120, 130) may select serving beams (112, 113, 121, 131) through a beam search or beam management procedure. After the serving beams (112, 113, 121, 131) are selected, subsequent communication may be performed through a resource that is in a quasi-co-located (QCL) relationship with the resource that transmitted the serving beams (112, 113, 121, 131).

[0049] The base station (110), the first terminal (120), and the second terminal (130) of the present disclosure may each be a transmitting apparatus, a transmitting node, a receiving apparatus, and / or a receiving node. For example, the base station (110) may transmit a radio frequency (RF) signal to the first terminal (120). The base station (110) may receive an RF signal from the first terminal (120). As another example, the first terminal (120) may transmit an RF signal to the base station (110) or the second terminal (130). The first terminal (120) may receive an RF signal from the base station (110) or the second terminal (130).

[0050] FIG. 2 is a drawing for explaining the structure of a terminal according to embodiments.

[0051] Referring to FIG. 2, a terminal (200) according to embodiments may include a transceiver (transmitter / receiver) (210), a memory (220), and / or a processor (230). Although the present disclosure describes the terminal (200) as including a transceiver (210), a memory (220), and / or a processor (230), this is merely an example. For example, the terminal (200) may include additional components other than the transceiver (210), the memory (220), and the processor (230).

[0052] According to the embodiments, the transceiver (210), memory (220), and processor (230) may each be implemented or formed as separate chips. However, this is merely an example, and the transceiver (210), memory (220), and / or processor (230) may be implemented or formed as a single chip.

[0053] According to embodiments, the transceiver (210) may include at least one transmitter and / or at least one receiver. For example, the transceiver (210) may include an RF transmitter for amplifying and up-converting the frequency of a transmitted signal. The transceiver (210) may include an RF receiver for down-converting the frequency of a received signal and amplifying low-noise.

[0054] The configurations of the transceiver (210) described in this disclosure are merely examples and the configuration of the transceiver (210) is not limited to an RF transmitter and an RF receiver. For example, the transceiver (210) may further include a coupler to ensure isolation between the RF transmitter and the RF receiver.

[0055] According to the embodiments, the transceiver (210) can transmit or receive a signal to or from the processor (230). For example, the transceiver (210) can transmit or deliver an RF signal received through a wireless communication channel to or from the processor (230). The transceiver (210) can receive or receive an RF signal from or from the processor (230).

[0056] According to the embodiments, the transceiver (210) may be referred to as a UE transmitter or a UE receiver.

[0057] According to embodiments, the transceiver (210) may transmit a signal to a base station (e.g., base station (110) of FIG. 1) or a network entity (e.g., an access and mobility management function (AMF) entity) or receive a signal from a base station or a network entity. In embodiments, the transmitted or received signal may include control signals and data.

[0058] According to embodiments, the memory (220) may include or store programs and data necessary for the operations of the terminal (200). For example, the memory (220) may be a non-transitory memory, and a program stored in the non-transitory memory may be organically coupled with the hardware configuration of the terminal (200) (e.g., a processor (230) or a transceiver (210)). The memory (220) may store control information or data including signals obtained by the terminal (200). In embodiments, the memory (220) may include a read-only memory (ROM), a random access memory (RAM), a hard disk, a CD-ROM, a DVD, and / or a storage medium.

[0059] According to the embodiments, the processor (230) may include one processor or a plurality of processors. For example, the processor (230) may include a communication processor. For example, the processor (230) may include a communication processor and / or an application processor.

[0060] According to embodiments, the processor (230) can control a series of processes performed by the terminal (200). For example, the transceiver (210) can receive a data signal containing control information transmitted by a base station or network entity. The processor (230) can process the received control signal and data signal.

[0061] The term processor in the present disclosure may be replaced with various terms referring to a configuration that executes or performs operations of the terminal (200). For example, the processor may be replaced with a controller or a computing circuit.

[0062] The terminal (200) of the present disclosure may correspond to the first terminal (120) and / or the second terminal (130) of FIG. 1.

[0063] FIG. 3 is a diagram illustrating the structure of a network entity (or base station) according to embodiments.

[0064] Referring to FIG. 3, a network entity (300) according to embodiments may include a transceiver (transmitter / receiver) (310), a memory (320), and / or a processor (330). Although the present disclosure describes the network entity (300) as including a transceiver (310), a memory (320), and / or a processor (330), this is merely an example. For example, the network entity (300) may include additional components other than the transceiver (310), the memory (320), and the processor (330). The network entity (300) may represent network functions included in a base station or other core network.

[0065] According to the embodiments, the transceiver (310), memory (320), and processor (330) may each be implemented or formed as separate chips. However, this is merely an example, and the transceiver (310), memory (320), and / or processor (330) may be implemented or formed as a single chip.

[0066] According to embodiments, the transceiver (310) may include at least one transmitter and / or at least one receiver. For example, the transceiver (310) may include an RF transmitter for amplifying and up-converting the frequency of a transmitted signal. The transceiver (310) may include an RF receiver for down-converting the frequency of a received signal and amplifying low-noise.

[0067] The configurations of the transceiver (310) described in this disclosure are merely examples and are not limited to an RF transmitter and an RF receiver. For example, the transceiver (310) may further include a coupler to ensure isolation between the RF transmitter and the RF receiver.

[0068] According to the embodiments, the transceiver (310) can transmit or receive a signal to or from the processor (330). For example, the transceiver (310) can transmit or deliver an RF signal received through a wireless communication channel to or from the processor (330). The transceiver (310) can receive or receive an RF signal from the processor (230).

[0069] According to the embodiments, the transceiver (310) may be referred to as a network entity transmitter or a network entity receiver.

[0070] According to embodiments, the transceiver (310) may transmit a signal to the terminal (200) or another network entity or receive a signal from the terminal (200) or another network entity. In embodiments, the transmitted or received signal may include control signals and data.

[0071] According to embodiments, the memory (320) may contain programs and data necessary for the operations of the network entity (300). For example, the memory (320) may be a non-transitory memory, and a program stored in the non-transitory memory may be organically coupled with the hardware configuration of the network entity (300) (e.g., a processor (330) or a transceiver (310)). The memory (320) may store control information or data including signals obtained by the network entity (300). In embodiments, the memory (320) may include read-only memory (ROM), random access memory (RAM), a hard disk, a CD-ROM, a DVD, and / or storage media.

[0072] According to the embodiments, the processor (330) may include one processor or a plurality of processors. For example, the processor (330) may include a communication processor. For example, the processor (330) may include a communication processor and / or an application processor.

[0073] According to embodiments, the processor (330) can control a series of processes performed by the network entity (300). For example, the transceiver (310) can receive a data signal containing control information transmitted by a terminal or another network entity. The processor (330) can process the received control signal and data signal.

[0074] The term processor in the present disclosure may be replaced with various terms referring to a configuration that executes or performs operations of a network entity (300). For example, processor may be replaced with a controller or a computing unit.

[0075] The network entity (300) of the present disclosure may correspond to the base station (110) of FIG. 1.

[0076] The device described in FIGS. 2 and 3 may correspond to a device of a transmitting end or a receiving end. A terminal or network entity according to embodiments of the present disclosure may be a transmitting end when it is a transmitting end, and may be a receiving end when the terminal or network entity is a receiving end.

[0077] The terminal, base station, or network entity described in FIGS. 1 to 3 can perform channel estimation based on SRS transmission and reception.

[0078] FIG. 4 illustrates an example of SRS transmission of a terminal and channel estimation of a base station according to embodiments of the present disclosure.

[0079] Referring to FIG. 4, terminals (UE_1, UE_2, ... UE_N) can transmit a sounding reference signal (SRS) to a base station (BS), and the base station can perform uplink channel estimation based on the received SRS. In addition, the base station can calculate a value for downlink beamforming (BF) based on the estimated channel, and beamform based on the calculated result to transmit data to the terminals through a downlink channel (e.g., physical downlink shared channel, PDSCH).

[0080] The terminal can receive configuration information related to SRS transmission (e.g., resources for SRS transmission, schedule information, etc.) from the base station and can transmit SRS based on the received configuration information. The base station can estimate the state of the channel based on the received SRS and calculate beamforming weights based on the estimated channel. Beamforming weights may relate to the phase and amplitude adjustment of the signal to be transmitted at each antenna element.

[0081] The method and apparatus according to the embodiments of the present disclosure relate to a technique for performing channel estimation using SRS. Specifically, the embodiments of the present disclosure may include a method in which auxiliary information is used for channel estimation, information related to previously received SRS is used, or an artificial intelligence model is used.

[0082] FIG. 5 illustrates an example of SRS reception and data generated in the antenna / frequency domain according to embodiments of the present disclosure.

[0083] Referring to FIG. 5, frequency and time resources for the transmission of SRS are exemplified. Orthogonal frequency division multiplexing (OFDM) symbols may be used for the transmission of SRS. As an example, SRS may be transmitted in a 2-comb form at the last symbol of a subframe (510).

[0084] A base station can receive SRS using multiple antennas (520). The base station can receive SRS symbols arranged in the frequency domain on an antenna-by-antenna basis to generate a matrix (530) of the dimension 'frequency domain × number of receiving antennas', and can generate a received signal matrix (540) of the dimension 'number of SRS elements × number of receiving antennas' by extracting only SRS elements from the frequency domain. At this time, the relationship between the received signal matrix and the channel matrix can be expressed as [Equation 1] below.

[0085] [Mathematical Formula 1]

[0086] Yp = Hp · Xp + Wp

[0087] Here, Yp represents the received signal matrix representing the SRS signal received at the base station, Xp is the matrix representing the SRS symbols known to the base station, and Hp can represent the channel matrix. Wp can represent channel noise. In Equation 1, the matrix product is an element-wise product, and it can be assumed that time and frequency offsets are fully compensated.

[0088] The embodiments of the present disclosure relate to a channel estimation method for deriving a more accurate Hp. The channel estimation method is based on statistics calculated from a received signal and may include the least squares (LS) method and the minimum mean square error (MMSE) method. The LS method has the advantage of being simple to calculate and having a low burden of complexity, but it may have relatively low performance due to sensitivity to noise. On the other hand, MMSE is a method that reflects the statistical characteristics of the channel and enables more accurate channel estimation compared to the LS method, but it requires statistical information on channel noise and has high computational complexity. That is, there exist methods with very high complexity (MMSE) and methods with low complexity (LS), and a channel estimation method capable of securing appropriate complexity and performance between them may be required.

[0089] FIG. 6 illustrates a method for estimating a channel using an artificial intelligence model according to embodiments of the present disclosure.

[0090] Referring to FIG. 6, an artificial intelligence model (620) may be used for channel estimation. As input to the artificial intelligence model (620), a channel matrix (610) estimated based on the LS method may be used. At this time, the estimated channel matrix (610) may include a real part and an imaginary part. The purpose of the artificial intelligence model (620) is to denoise as much as possible the noise elements included in the channel matrix (610). The artificial intelligence model (620) may be trained using a noise-free channel matrix as label data (or ground truth data). Thus, the artificial intelligence model (620) may receive the estimated channel matrix (610) containing noise as input and output a channel matrix (630) in which the noise has been denoised as much as possible.

[0091] Such artificial intelligence models may include various models for noise removal in input data, such as models based on convolutional neural networks (CNN), models based on recurrent neural networks (RNN), and models based on generative adversarial networks (GAN).

[0092] Channel estimation methods using artificial intelligence models can have lower complexity while achieving higher performance than LMMSE (linear minimum mean square error).

[0093] Embodiments of the present disclosure may utilize additional auxiliary information for channel estimation. For example, the accuracy of channel estimation may be improved based on information regarding a channel received from a terminal or information that the base station can independently obtain or estimate. The auxiliary information may be utilized in various ways, such as being used as input for an artificial intelligence model or for determining whether to use previously estimated channel estimation results for current channel estimation. When past channel estimation results are utilized, changes in the channel or environment over time can be taken into account, thereby improving the performance of channel estimation.

[0094] In addition, auxiliary information can enable the use of artificial intelligence models specialized for base stations. For example, when using methods such as LS and MMSE, each model must be installed in common on all base stations, but when using artificial intelligence models, a model reflecting the environment of each base station can be installed on the base station, and since continuous learning of the model is possible, channel estimation that accurately reflects the characteristics of each base station can be performed.

[0095] FIGS. 7A and 7B illustrate a method for estimating a channel using an artificial intelligence model according to embodiments of the present disclosure.

[0096] Specifically, FIGS. 7A and 7B exemplarily illustrate an artificial intelligence model that receives a channel estimation matrix derived in the past or a received signal matrix acquired in the past as input and outputs a current channel estimation matrix.

[0097] Referring to FIG. 7A, the received signal matrix Yp(t) generated based on the SRS received at time t, the received signal matrix Yp(t-1) generated based on the SRS received at time t-1, and the received signal matrix Yp(t-2) generated based on the SRS received at time t-2 can be input into the first artificial intelligence model and used for channel estimation. That is, current channel estimation can be performed based on the SRS received in the past.

[0098] The first artificial intelligence model can receive Yp(t) (712) as input and output a first channel estimation matrix (722), receive Yp(t-1) (714) as input and output a second channel estimation matrix (724), and receive Yp(t-2) (716) as input and output a third channel estimation matrix (726). The second channel estimation matrix (724) and the third channel estimation matrix (726) may be data that the base station has previously stored in memory.

[0099] The second artificial intelligence model can receive the first channel estimation matrix (722), the second channel estimation matrix (724), and the third channel estimation matrix (726) as inputs and output a final channel estimation matrix (730). That is, the second artificial intelligence model can use channel estimation matrices (724, 726) based on past SRS (t-1, t-2, etc.) as additional inputs. Additionally, the second artificial intelligence model can use auxiliary information for channel estimation as inputs to output a final channel estimation matrix (730) with improved accuracy. Since the second artificial intelligence model receives multiple channel estimation matrices as inputs, it may be referred to as a synthesis model.

[0100] In one embodiment, when inputting a plurality of channel estimation matrices into a second artificial intelligence model, various methods may be utilized. For example, a method of inputting each channel estimation matrix by summing (or weighted summing), a method of inputting each channel estimation matrix by concatenating them, or a method of inputting channel estimation matrices by combining them based on arbitrary statistics (e.g., Norm, Cosine similarity, etc.) for each channel estimation matrix may be applied. The method of inputting a plurality of channel estimation matrices is not limited to the examples described above, and various methods may be considered within the scope that can be derived by a person skilled in the art.

[0101] In the case of Fig. 7A, a previously stored past channel estimation matrix can be optionally used, and implementation can be achieved without a significant increase in complexity.

[0102] Referring to FIG. 7B, the received signal matrix Yp(t) (712), generated based on the SRS received at time t, the received signal matrix Yp(t-1) (714), generated based on the SRS received at time t-1, and the received signal matrix Yp(t-2) (716), generated based on the SRS received at time t-2, can be used for channel estimation. That is, current channel estimation can be performed based on SRS received in the past.

[0103] In FIG. 7B, the third artificial intelligence model can output a feature vector for each received signal matrix. Then, the fourth artificial intelligence model can receive each of the output feature vectors as input and output a final channel estimation matrix (750). The feature vectors output by the third artificial intelligence model can be understood as compressed information containing features of the received signal matrix, and generally may be a latent vector consisting of a single line with a size much smaller than the input.

[0104] The third artificial intelligence model can receive Yp(t) (712) as input and output a first feature vector (742), receive Yp(t-1) (714) as input and output a second feature vector (744), and receive Yp(t-2) (716) as input and output a third feature vector (746). The third artificial intelligence model may be referred to as a feature extraction model.

[0105] The fourth artificial intelligence model can receive the first feature vector (742), the second feature vector (744), and the third feature vector (746) as inputs and output a final channel estimation matrix (750). That is, the fourth artificial intelligence model can use the feature vectors of the received signal matrix based on past SRS as inputs. Additionally, the fourth artificial intelligence model can use auxiliary information for channel estimation as inputs to output a final channel estimation matrix (750) with improved accuracy. The fourth artificial intelligence model may be referred to as an estimation model.

[0106] In the case of Fig. 7B, the training process of the model can be simplified because the training is performed based on a feature vector that includes the features of the received signal matrix but is small in size.

[0107] In various embodiments, slots, subframes, etc., including past SRSs considered for channel estimation may be selected based on auxiliary information. That is, some of the various time points such as t-1, t-2, t-3, ..., tn may be considered for channel estimation based on auxiliary information.

[0108] FIG. 8 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0109] Referring to FIG. 8, an embodiment is described in which a terminal transmits auxiliary information for channel estimation to a base station, and the base station performs channel estimation based on the received auxiliary information.

[0110] In an 812 operation, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include frequency resources, time resources, scheduling information, or information regarding SRS transmission power for SRS transmission.

[0111] In the 814 operation, the terminal can transmit SRS to the base station based on configuration information received from the base station.

[0112] In operation 816, the base station may perform channel estimation based on the SRS received from the terminal. With respect to channel estimation, the channel estimation method described in FIGS. 4 to 7 and FIGS. 13A to 15 may be applied.

[0113] In an 818 operation, the terminal may transmit auxiliary information for channel estimation to the base station. The auxiliary information for channel estimation may include CQI (channel quality information), RSRP (reference signal received power), RSRQ (reference signal received quality), RSSI (reference signal strength indicator), SINR (signal to interference plus noise ratio), SNR (signal to noise ratio), I / N (interference to noise ratio), Doppler shift, Delay spread, the speed of the terminal, or various other measurements taken by the terminal's sensors.

[0114] CQI can represent the downlink channel quality measured by the terminal.

[0115] RSRP can represent the value of the received power of the reference signal of a specific cell.

[0116] RSSI can represent the power of the total received signal.

[0117] RSRQ can represent a reception quality indicator based on RSRP and RSSI.

[0118] SINR represents the ratio of signal to interference and noise, and can indicate the actual quality of the channel.

[0119] SNR represents the signal-to-noise ratio, and I / N can represent the interference-to-noise ratio.

[0120] Doppler shift represents the frequency change based on the relative speed between the terminal and the base station, and delay spread can represent the dispersion of the time it takes for multipath signals to arrive.

[0121] The aforementioned auxiliary information may be periodically transmitted from the terminal to the base station, and the most recent data from the time of channel estimation may be transmitted.

[0122] In operation 820, the base station may set a channel estimation method based on received auxiliary information. For example, the base station may use the received auxiliary information as input to an artificial intelligence model for channel estimation (e.g., the second artificial intelligence model of FIG. 7A), or select a past received signal matrix or a past channel estimation matrix to use as input to the artificial intelligence model using the received auxiliary information. Alternatively, the base station may select an artificial intelligence model to use for channel estimation among a plurality of artificial intelligence models using the received auxiliary information.

[0123] In one embodiment, the base station may determine that the quality of the channel is poor based on received auxiliary information (e.g., CQI or RSRQ) and may consider the channel change pattern by utilizing a larger number of past SRS reception information or past channel estimation information. Alternatively, if the base station determines that the quality of the channel is good based on the received auxiliary information, that is, if it determines that channel estimation is sufficient with only the currently received SRS, the base station may prevent the artificial intelligence model from using past data as input.

[0124] In an 822 operation, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for SRS transmission. In this case, the configuration information may be based on the channel estimation method set in the 820 operation. Therefore, the contents of the configuration information in the 812 operation and the configuration information in the 822 operation may be different.

[0125] In operation 824, the terminal can transmit SRS to the base station based on the configuration information received in operation 822.

[0126] In operation 826, the base station can perform channel estimation based on the channel estimation method set in operation 820 and the SRS received in operation 824.

[0127] In the embodiment of FIG. 8, some operations may be omitted or their order changed, and multiple operations may be expressed in an integrated manner. That is, the embodiments of the present disclosure may include ordinary variations of the embodiment of FIG. 8. Furthermore, the embodiment of FIG. 8 may be combined with technical ideas described in other drawings to form new embodiments, and such combination falls within the scope of the embodiments of the present disclosure.

[0128] FIG. 9 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0129] Referring to FIG. 9, an embodiment is described in which a terminal generates an indicator regarding a channel estimation method based on auxiliary information for channel estimation and transmits the generated indicator to a base station.

[0130] In a scenario designed to reduce the complexity of an artificial intelligence model operated at a base station, the terminal can generate an indicator to specify the base station's channel estimation method based on auxiliary information and transmit the generated indicator to the base station. That is, the terminal can aggregate and analyze various available auxiliary information, map the results to an indicator composed of one bit or multiple bits, and transmit it to the base station. The method by which the terminal synthesizes auxiliary information can vary, and it may follow criteria predefined with the base station or use a separate artificial intelligence model within the terminal.

[0131] In a 912 operation, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for SRS transmission.

[0132] In 914 operation, the terminal can transmit SRS to the base station based on configuration information received from the base station.

[0133] In operation 916, the base station may perform channel estimation based on the SRS received from the terminal. With respect to channel estimation, the channel estimation method described in FIGS. 4 to 7 and FIGS. 13A to 15 may be applied.

[0134] In the 918 operation, the terminal may generate an indicator regarding the channel estimation method based on auxiliary information for channel estimation. Auxiliary information for channel estimation may include CQI (channel quality information), RSRP (reference signal received power), RSRQ (reference signal received quality), RSSI (reference signal strength indicator), SINR (signal to interference plus noise ratio), SNR (signal to noise ratio), I / N (interference to noise ratio), Doppler shift, delay spread, the speed of the terminal, or various other measurements taken by the terminal's sensors. The aforementioned auxiliary information may be information periodically transmitted from the terminal to the base station, and may be the most recent data from the time of channel estimation.

[0135] The terminal may generate an indicator to indicate the channel estimation method of the base station, such as indicating how many past SRS received signal matrices or past channel estimation matrices to use for channel estimation by considering auxiliary information such as CQI, RSRQ, and the terminal's rate information, or indicating the weighting ratio for past data. For example, the terminal may generate an indicator composed of one bit or multiple bits to indicate the channel estimation method of the base station.

[0136] In various embodiments, an indicator for indicating a channel estimation method of a base station may include an indicator for indicating various elements regarding the channel estimation method, such as the type of artificial intelligence model that the base station must use for channel estimation, whether to use past data as input to the artificial intelligence model, how many past data must be used as input to the artificial intelligence model, and what data must be used as input.

[0137] In various embodiments, if the terminal determines that the quality of the channel is poor based on acquired auxiliary information (e.g., CQI or RSRQ), it may generate an indicator so that the base station can consider the channel change pattern by utilizing a larger number of past SRS reception information or past channel estimation information. Alternatively, if the terminal determines that the quality of the channel is good based on the received auxiliary information, that is, if it determines that channel estimation is sufficient with only the currently received SRS, the terminal may generate an indicator so that the base station's artificial intelligence model does not use past data as input.

[0138] In operation 922, the base station may set the channel estimation method based on an indicator regarding the channel estimation method received from the terminal. For example, the base station may use the received auxiliary information as input to an artificial intelligence model for channel estimation according to the content indicated by the indicator, or the base station may select a past received signal matrix or a past channel estimation matrix to use as input to the artificial intelligence model. Alternatively, the base station may select an artificial intelligence model to use for channel estimation from among a plurality of artificial intelligence models based on the indicator.

[0139] For example, when a base station receives a first indicator, the base station may perform channel estimation using only the currently received SRS as input to an artificial intelligence model. Alternatively, when a base station receives a second indicator (different from the first indicator), the base station may perform channel estimation using the received signal matrix or channel estimation matrix generated by the currently received SRS as well as the previously received SRS as input to an artificial intelligence model.

[0140] In operation 924, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for SRS transmission. In this case, the configuration information may be based on the channel estimation method set in operation 922. Therefore, the contents of the configuration information in operation 912 and the configuration information in operation 924 may be different.

[0141] In operation 926, the terminal can transmit SRS to the base station based on the configuration information received in operation 924.

[0142] In operation 928, the base station can perform channel estimation based on the channel estimation method set in operation 922 and the SRS received in operation 926.

[0143] In the embodiment of FIG. 9, some operations may be omitted or their order changed, and multiple operations may be expressed in an integrated manner. That is, the embodiments of the present disclosure may include ordinary variations of the embodiment of FIG. 9. Furthermore, the embodiment of FIG. 9 may be combined with technical ideas described in other drawings to form new embodiments, and such combination falls within the scope of the embodiments of the present disclosure.

[0144] FIG. 10 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0145] Referring to FIG. 10, an embodiment is described in which a base station receives auxiliary information for channel estimation from a terminal, and the base station selects an artificial intelligence model for channel estimation based on the auxiliary information.

[0146] A base station may store multiple candidate AI models for channel estimation. Each candidate AI model stored in the base station may be optimized based on the quality or other characteristics of the received signal. Since the signals received by the base station may possess specific characteristics depending on the environment in which the base station is installed, the candidate AI models may be trained to improve channel estimation performance for signals with specific characteristics. By determining the signal transmission environment based on auxiliary information and selecting one of the stored candidate AI models to perform channel estimation, the performance of channel estimation can be improved even if the installation environments of each base station differ.

[0147] In operation 1012, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for SRS transmission.

[0148] In operation 1014, the terminal can transmit SRS to the base station based on configuration information received from the base station.

[0149] In operation 1016, the base station may perform channel estimation based on the SRS received from the terminal. With respect to channel estimation, the channel estimation method described in FIGS. 4 to 7 and FIGS. 13A to 15 may be applied.

[0150] In operation 1018, the terminal may transmit auxiliary information for channel estimation to the base station. The auxiliary information for channel estimation may include CQI (channel quality information), RSRP (reference signal received power), RSRQ (reference signal received quality), RSSI (reference signal strength indicator), SINR (signal to interference plus noise ratio), SNR (signal to noise ratio), I / N (interference to noise ratio), Doppler shift, Delay spread, the speed of the terminal, or various other measurements taken by the terminal's sensors. The aforementioned auxiliary information may be transmitted periodically from the terminal to the base station, and the most recent data from the time of channel estimation may be transmitted.

[0151] In operation 1020, the base station may determine an artificial intelligence model to be used for channel estimation among a plurality of candidate artificial intelligence models based on auxiliary information received in operation 1018. At this time, an auxiliary information processing model that receives auxiliary information as input and outputs an index indicating a channel estimation method may be used. For example, the base station may input auxiliary information into the auxiliary information processing model and use an artificial intelligence model corresponding to the output index for channel estimation. Here, the artificial intelligence model corresponding to the output index may output a channel estimation matrix by receiving an SRS received at one point in time as input, as in the first artificial intelligence model of FIG. 7A, or output a channel estimation matrix by receiving an SRS received at multiple points in time, including past points in time, as input. The configuration of the artificial intelligence model and the input / output information are not limited to those described in this disclosure and may vary.

[0152] In one embodiment, the output index may correspond to a single artificial intelligence model or to a specific channel estimation method. For example, the channel estimation method corresponding to the output index may represent a method of generating a final channel estimation matrix by inputting a channel estimation matrix based on the past SRS and a channel estimation matrix based on the current SRS into a specific artificial intelligence model. Various channel estimation methods may be mapped to the index.

[0153] In one embodiment, the auxiliary information processing model may be learned according to the characteristics of terminals or installation environments generally experienced by the base station, and different models may be stored for each base station. Therefore, since different environments for each base station are taken into account in channel estimation, channel estimation performance may be improved.

[0154] In one embodiment, when the performance of the base station is insufficient or computing resources are insufficient due to a large number of terminals connecting simultaneously, only the channel estimation model may be used without using the aggregation model, as directed by an auxiliary information processing model.

[0155] In operation 1022, the base station may set a channel estimation method based on received auxiliary information and / or a determined artificial intelligence model. For example, the base station may use the received auxiliary information as input to an artificial intelligence model for channel estimation (e.g., the second artificial intelligence model of FIG. 7A), or select a past received signal matrix or a past channel estimation matrix to use as input to the artificial intelligence model using the received auxiliary information. Additionally, the base station may set a channel estimation method associated with an artificial intelligence model corresponding to the index output in operation 1020. The channel estimation method may be defined by the index output in operation 1020.

[0156] In one embodiment, the auxiliary information may be used only as input information for selecting an artificial intelligence model and may not be used as input to a collection model for channel estimation. That is, the auxiliary information may be used only for constructing an artificial intelligence model for channel estimation.

[0157] In one embodiment, if auxiliary information (e.g., CQI or RSRQ) indicates that the channel quality is poor, the auxiliary information processing model may output an index of a model that can consider the channel change pattern by utilizing a larger number of past SRS reception information or past channel estimation information. Alternatively, if the auxiliary information indicates that the channel quality is good, the auxiliary information processing model may output an index of a model that does not utilize past data as input.

[0158] Operations 1020 and 1022 are separated for convenience of explanation, but they can be represented in a single block in a flowchart. That is, operations 1020 and 1022 may be performed together.

[0159] In operation 1024, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for SRS transmission. In this case, the configuration information may be based on the channel estimation method set in operation 1022. Therefore, the contents of the configuration information in operation 1012 and the configuration information in operation 1024 may be different.

[0160] In operation 1026, the terminal can transmit SRS to the base station based on the configuration information received in operation 1024.

[0161] In operation 1028, the base station can perform channel estimation based on the artificial intelligence model determined in operation 1020, the channel estimation method set in operation 1022, and / or the SRS received in operation 1024.

[0162] In the embodiment of FIG. 10, some operations may be omitted or their order changed, and multiple operations may be expressed in an integrated manner. That is, the embodiments of the present disclosure may include ordinary variations of the embodiment of FIG. 10. Furthermore, the embodiment of FIG. 10 may be combined with the technical concept described in other drawings to form a new embodiment, and such combination falls within the scope of the embodiments of the present disclosure.

[0163] FIG. 11 illustrates a method for estimating a channel according to embodiments of the present disclosure.

[0164] Referring to FIG. 11, an embodiment is described in which a base station detects an anomaly in an SRS received from a terminal, requests auxiliary information (or channel quality information) for channel estimation from the terminal, receives auxiliary information from the terminal, and selects an artificial intelligence model for channel estimation based on the received auxiliary information.

[0165] A base station may store multiple candidate AI models for channel estimation. Each candidate AI model stored in the base station may be a model optimized according to the quality or other characteristics of the received signal.

[0166] In operation 1112, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for SRS transmission.

[0167] In operation 1114, the terminal can transmit SRS to the base station based on configuration information received from the base station.

[0168] In operation 1116, the base station can perform channel estimation based on the SRS received from the terminal. With respect to channel estimation, the channel estimation method described in FIGS. 4 to 7 and FIGS. 13A to 15 may be applied. At this time, the base station can detect outliers in the received SRS. Outlier detection in the SRS may indicate cases where the difference between the currently received SRS and the SRSs that the base station has normally received is excessive.

[0169] In operation 1118, the base station may request auxiliary information for channel estimation from the terminal. Operation 1118 may be performed when the base station detects an anomaly in the SRS in operation 1116. The request from the base station may include the type of auxiliary information required.

[0170] In operation 1120, the terminal may transmit auxiliary information for channel estimation to the base station in response to the request of operation 1118. The auxiliary information for channel estimation may include CQI (channel quality information), RSRP (reference signal received power), RSRQ (reference signal received quality), RSSI (reference signal strength indicator), SINR (signal to interference plus noise ratio), SNR (signal to noise ratio), I / N (interference to noise ratio), Doppler shift, Delay spread, the speed of the terminal, or various other measurements taken by the terminal's sensors. The aforementioned auxiliary information may be transmitted periodically from the terminal to the base station, and the most recent data from the time of channel estimation may be transmitted.

[0171] In operation 1122, the base station may determine an artificial intelligence model to be used for channel estimation among a plurality of candidate artificial intelligence models based on auxiliary information received in operation 1120. At this time, an auxiliary information processing model that receives auxiliary information as input and outputs an index representing a channel estimation method may be used. For example, the base station may input auxiliary information into the auxiliary information processing model and use an artificial intelligence model corresponding to the output index for channel estimation. Here, the artificial intelligence model corresponding to the output index may output a channel estimation matrix by receiving an SRS received at one point in time as input, as in the first artificial intelligence model of FIG. 7A, or it may output a channel estimation matrix by receiving an SRS received at multiple points in time, including past points in time, as input. The configuration of the artificial intelligence model and the input / output information are not limited to those described in this disclosure and may vary.

[0172] In one embodiment, the output index may correspond to a single artificial intelligence model or to a specific channel estimation method. For example, the channel estimation method corresponding to the output index may represent a method of generating a final channel estimation matrix by inputting a channel estimation matrix based on the past SRS and a channel estimation matrix based on the current SRS into a specific artificial intelligence model. Various channel estimation methods may be mapped to the index.

[0173] In one embodiment, the auxiliary information processing model may be learned according to the characteristics of terminals or installation environments generally experienced by the base station, and different models may be stored for each base station. Therefore, since different environments for each base station are taken into account in channel estimation, channel estimation performance may be improved.

[0174] In operation 1124, the base station may set a channel estimation method based on received auxiliary information and / or a determined artificial intelligence model. For example, the base station may use the received auxiliary information as input to an artificial intelligence model for channel estimation (e.g., the second artificial intelligence model in FIG. 7A), or select a past received signal matrix or a past channel estimation matrix to use as input to the artificial intelligence model using the received auxiliary information. Additionally, the base station may set a channel estimation method associated with an artificial intelligence model corresponding to the index output in operation 1122. The channel estimation method may be defined by the index output in operation 1122.

[0175] In one embodiment, the auxiliary information may be used only as input information for the selection (or determination) of an artificial intelligence model and may not be used as input to a collection model for channel estimation. That is, the auxiliary information may be used only for the configuration of an artificial intelligence model for channel estimation.

[0176] In one embodiment, if auxiliary information (e.g., CQI or RSRQ) indicates that the channel quality is poor, the auxiliary information processing model may output an index of a model that can consider the channel change pattern by utilizing a larger number of past SRS reception information or past channel estimation information. Alternatively, if the auxiliary information indicates that the channel quality is good, the auxiliary information processing model may output an index of a model that does not utilize past data as input.

[0177] Actions 1122 and 1124 are separated for convenience of explanation, but they can be represented in a single block in a flowchart. That is, actions 1122 and 1124 may be performed together.

[0178] In operation 1126, the base station may transmit configuration information related to SRS transmission to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for SRS transmission. In this case, the configuration information may be based on the channel estimation method set in operation 1124. Therefore, the contents of the configuration information in operation 1112 and the configuration information in operation 1126 may be different.

[0179] In operation 1128, the terminal can transmit SRS to the base station based on the configuration information received in operation 1126.

[0180] In operation 1130, the base station can perform channel estimation based on the artificial intelligence model determined in operation 1122, the channel estimation method set in operation 1124, and / or the SRS received in operation 1128.

[0181] In the embodiment of FIG. 11, some operations may be omitted or their order changed, and multiple operations may be expressed in an integrated manner. That is, the embodiments of the present disclosure may include ordinary variations of the embodiment of FIG. 11. Furthermore, the embodiment of FIG. 11 may be combined with the technical concept described in other drawings to form a new embodiment, and such combination falls within the scope of the embodiments of the present disclosure.

[0182] FIGS. 12A and FIGS. 12B illustrate a method for estimating a channel according to embodiments of the present disclosure.

[0183] Referring to Fig. 12A, the process of a base station learning an anomaly detection model for anomaly detection of SRS and detecting anomalies using the anomaly detection model is explained.

[0184] In FIG. 12A, the base station is divided into a CU (central unit) and a DU (distributed unit), and the training of the outlier detection model and outlier detection are shown to be performed in the CU, but is not limited thereto and can also be performed in the DU.

[0185] 1220 may represent a series of channel estimation cycles in which the base station receives auxiliary information from the terminal, sets a channel estimation method based on the received auxiliary information, and performs channel estimation by receiving the SRS.

[0186] Channel estimation cycles such as 1220 are performed repeatedly, and during this process, the base station can obtain information about the results of the channel estimation from the terminal. That is, the base station can know information about whether the previous channel estimation results accurately represented the state of the actual channel.

[0187] As in operation 1222, the base station (CU or DU) can receive information about the channel estimation result from the terminal.

[0188] In operation 1224, the base station (DU) can train an outlier detection model by configuring training data based on information regarding channel estimation results. In one embodiment, the training data for training the outlier detection model may be configured by selecting cases where the channel estimation results were good and cases where the results were bad from the information regarding channel estimation results, and configuring the SRS received signal matrix associated with each case as training data.

[0189] In one embodiment, the outlier detection model may include an autoencoder model comprising an encoder and a decoder. Accordingly, the outlier detection model can output a semantically identical matrix by receiving an SRS received signal matrix as input, encoding and compressing it, and decoding and restoring it. Here, whether the SRS is an outlier can be determined based on the difference between the input and the output. For example, when the difference between the input and the output is above or above a specific threshold, the base station can identify that the received SRS corresponds to an outlier. When the difference between the input and the output is below or below the threshold, the base station can identify that the received SRS does not correspond to an outlier.

[0190] In one embodiment, the outlier detection model may be a model trained on labeled training data. The labeled training data may include data indicating whether the channel estimation result is good or bad for the received SRS. Accordingly, the outlier detection model may be trained on SRSs with good or bad channel estimation results, and may take an SRS received signal matrix as input and output a result indicating whether it is an outlier.

[0191] In operation 1226, the CU can transmit information about SRS detected as an outlier to the DU. That is, the base station can identify SRS outliers. The base station can identify SRS corresponding to outliers among the SRSs received in the past.

[0192] After the base station identifies an SRS outlier, in the next channel estimation cycle, the base station may establish a channel estimation method based on the SRS outlier information and perform channel estimation (1230). For example, the base station may not use information related to the SRS outlier when performing channel estimation. Specifically, the base station may not use the received signal matrix for past received SRSs identified as outliers and the channel estimation matrix based thereon in the current channel estimation. Thus, the base station may consider the characteristics of the SRS received signal according to the installation location and other surrounding environment, and improve the accuracy of channel estimation by not considering information regarding outlier SRSs that deviate from the normal range.

[0193] FIG. 12B illustrates the channel estimation operation (1230) of FIG. 12A. As described above, the base station may not consider SRS outliers in channel estimation. Specifically, if the received signal matrix Yp(t-1) at time t-1 corresponds to an SRS outlier, Yp(t-1) cannot be used as an input to an artificial intelligence model for channel estimation. FIG. 12B illustrates a process in which some of the received signal matrices from past time points are not input to the artificial intelligence model based on SRS outlier information. Specifically, a drop gate can be controlled so that data regarding SRS outliers among the past data is not input to the artificial intelligence model.

[0194] FIGS. 13A and FIGS. 13B illustrate a method for estimating a channel considering a Doppler shift according to embodiments of the present disclosure.

[0195] Referring to Fig. 13A, a base station can utilize auxiliary information measurable at the base station for channel estimation. For example, the demodulation reference signal (DM-RS) is a type of reference signal for channel estimation, and DM-RS and SRS located in the same band have a high correlation with each other. Therefore, the base station can add the previous Doppler shift information, which can be estimated from the DM-RS, as input to an artificial intelligence model for channel estimation to learn the correlation between the past SRS and the current SRS. Depending on the base station environment, the base station may utilize more complex indicators, such as the correlation matrix between DM-RS, in addition to the Doppler shift measurable from the DM-RS.

[0196] Referring to FIG. 13A, the base station can generate a Doppler shift from the DM-RS received from the terminal. Then, the Doppler shift can be added as an input to the second artificial intelligence model (e.g., a synthesis model) described in FIG. 7a. That is, the second artificial intelligence model can receive the Doppler shift as an input as well as the channel estimation matrix for the past SRS received signal matrix and output a final channel estimation matrix.

[0197] Referring to FIG. 13B, a base station can adjust the input of an artificial intelligence model for channel estimation based on a measured Doppler shift. For example, if the value of the measured Doppler shift exceeds a specific threshold, the base station may not utilize data related to the slot in which the Doppler shift was measured. Specifically, it may not utilize SRS-related data located in the symbols of the slot containing the DM-RS associated with the Doppler shift for channel estimation. Thus, if the measured Doppler shift in the slot received at time t-2 exceeds a specific threshold, the base station may not input the SRS received signal matrix Yp(t-2) received at time t-2 into the channel estimation model. Specifically, a drop gate may receive and determine the Doppler shift to filter past data to be used as input to the artificial intelligence model.

[0198] In various embodiments, in addition to Doppler shift, various information that can be measured at the base station can be utilized as described above.

[0199] FIG. 13A or FIG. 13B can be applied in combination with the channel estimation method described in other figures.

[0200] FIG. 14 illustrates an example of the operation of an artificial intelligence model for estimating a channel according to embodiments of the present disclosure.

[0201] Referring to FIG. 14, an example of how an artificial intelligence model (1410) operates recursively is described. For example, the artificial intelligence model (1410) can take the channel estimation matrix Hp(t-2) at time t-2 and the SRS received signal matrix Yp(t-1) at time t-1 as inputs and output the channel estimation matrix Hp(t-1) at time t-1. Also, the artificial intelligence model (1410) can take the channel estimation matrix Hp(t-1) at time t-1 and the SRS received signal matrix Yp(t) at time t as inputs and output the channel estimation matrix Hp(t) at time t.

[0202] In this way, when data for each time point is recursively input into the artificial intelligence model (1410), the earlier the data is input, the more it is diluted by the data input later, so relatively unimportant data can be learned with a low weight.

[0203] In the case of the artificial intelligence model (1410) described in Fig. 14, the channel estimation method can be adjusted so that a large number of past time points of data are recursively input when the correlation between the past SRS and the current SRS is high, and a small number of past time points of data are input when the correlation between the past SRS and the current SRS is low.

[0204] In various embodiments, the operation method of the artificial intelligence model (1410) of FIG. 14 can be applied to the channel estimation operation described in FIG. 8 to FIG. 12A. That is, when considering past SRS during channel estimation, data regarding past SRS can be recursively input into the artificial intelligence model for channel estimation.

[0205] FIG. 15 illustrates an example of the operation of an artificial intelligence model for estimating a channel according to embodiments of the present disclosure.

[0206] Referring to FIG. 15, the form of an artificial intelligence model (1500) capable of hierarchically inputting past data is described. The artificial intelligence model (1500) of FIG. 15 can reduce complexity while reflecting information about the flow of time.

[0207] The artificial intelligence model (1500) may include a layer that receives data and a drop layer that determines whether to use the input data. For example, the artificial intelligence model (1500) may include a first input layer (1512) that receives the received signal matrix Yp(t-2) at time t-2 as input and a first drop layer (1514) that determines whether to use Yp(t-2), a second input layer (add layer) (1516) that receives the received signal matrix Yp(t-1) at time t-1 as input, and a second drop layer (1518) that determines whether to use the input. In this case, the second input layer may add Yp(t-1) to the existing input data and reflect it (other operations other than addition are possible). Therefore, if Yp(t-2) is not dropped in the first drop layer, Yp(t-1) input in the second input layer can be added to Yp(t-2), and if Yp(t-2) is dropped in the first drop layer, Yp(t-1) can be added to the zero matrix. Then, the second drop layer (1518) can retain or drop the input to which Yp(t-1) has been added. Additionally, the artificial intelligence model (1500) includes a third input layer (1520) that receives Yp(t) at the current time point and can add Yp(t) to the existing input. At this time, the existing input is the input that was dropped or retained by the second input layer (1518). The artificial intelligence model (1500) further includes a channel estimation layer (1522), and the input that has passed through the third input layer (1520) is input to the channel estimation layer (1522) so that a final channel estimation matrix can be output.

[0208] Through such hierarchical input, historical data can be processed in fewer layers, and by ensuring that older data is input first at earlier layers, information of lower importance can be naturally diluted. Additionally, if data from a specific past point in time is deemed unnecessary, it can be discarded by the drop layer.

[0209] In various embodiments, the drop layer may be implemented as a successive decision-maker that determines how an input is reflected in the next input by applying weights to an input received at a specific layer.

[0210] In various embodiments, if there is a concern that the drop layer may cause excessive information deletion, a form may be configured such that information about at least one previous input layer is transmitted in a feed-forward format after the output of the drop layer. For example, to prevent a situation where information from the first input layer (1512) is useful and transmitted to a layer after the first drop layer (1514), but is completely removed along with the information from the second input layer (1516) at the second drop layer (1518), the output of the first drop layer (1514) may be connected to and processed with the output of the second drop layer (1518). Accordingly, the information of the first input layer (1512) may be processed directly with the output of the second drop layer (1518).

[0211] In various embodiments, the width of the feed-forward format may be one or more, and the structure of the feed-forward format may vary. For example, information regarding a first input layer may be connected to the output of another input layer beyond at least one input layer. Additionally, for example, information with weights applied to all or part of the information regarding the first input layer may be connected to the output of another input layer.

[0212] In various embodiments, the operation method of the artificial intelligence model (1500) of FIG. 15 can be applied to the channel estimation operation described in FIG. 8 to FIG. 12A. That is, when considering past SRS during channel estimation, data regarding past SRS can be hierarchically input into the artificial intelligence model for channel estimation.

[0213] In various embodiments, the number of input layers and drop layers can be adjusted.

[0214] Figure 15 can be applied in combination with the channel estimation method described in other figures.

[0215] FIG. 16 illustrates an example of the operation of a memory for estimating a channel according to embodiments of the present disclosure.

[0216] Referring to FIG. 16, an example of how a base station operates memory to reference past data in channel estimation is described. Specifically, the memory can store past data in order. For example, Yp(t), Yp(t-1), and Yp(t-2) can be stored sequentially in the memory storage space. The base station can store past data in a specific number (Tmax) of memory spaces according to the settings. However, the memory space set for storing past data may change depending on the channel estimation method. For example, if the number of past data referenced differs depending on the channel estimation method, the memory settings for storing past data may be changed accordingly.

[0217] Furthermore, when the base station acquires a new Yp(t+1), it can remove the data that was entered first from memory and store the newly acquired Yp(t+1) in the freed-up space. This method can be referred to as a first-in-first-out (FIFO) form. In other words, the data stored first can be deleted first.

[0218] In addition, if some past data has been deleted from the base station's memory or if there is no past data, the space in the memory where there is no past data may be filled with zeros.

[0219] FIG. 17 illustrates an example of the operation of a base station for estimating a channel according to embodiments of the present disclosure.

[0220] Referring to FIG. 17, in operation 1712, the base station transmits configuration information for SRS to the terminal. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for the terminal's SRS transmission.

[0221] In operation 1714, the base station can receive SRS from the terminal.

[0222] In operation 1716, the base station can perform channel estimation for wireless communication using an artificial intelligence model based on the received SRS.

[0223] An AI model for channel estimation may include a CNN-based AI model for denoising. Additionally, an AI model for channel estimation may include various models for denoising.

[0224] Channel estimation according to one embodiment may utilize a received signal matrix generated by a base station based on SRS received in the past, or a channel estimation matrix estimated by said received signal matrix. For example, the base station may use a first artificial intelligence model to calculate a first channel estimation matrix for the current SRS and a second channel estimation matrix for past SRS, and input the first channel estimation matrix and the second channel estimation matrix into a second artificial intelligence model to derive the current final channel estimation matrix. The second artificial intelligence model may utilize an artificial intelligence-based aggregation model that receives a plurality of channel estimation matrices as input.

[0225] In one embodiment, the base station can extract feature vectors from the current received signal matrix and the past received signal matrix, respectively, using a third artificial intelligence model. Then, the base station can derive a final channel estimation matrix by inputting each extracted feature vector into a fourth artificial intelligence model.

[0226] In one embodiment, a base station can derive a final channel estimation matrix by recursively inputting channel estimation matrices for past SRSs into an artificial intelligence model. Specifically, the base station can obtain a channel estimation matrix at time t-1 by inputting the channel estimation matrix at time t-2 and the received signal matrix at time t-1 into the artificial intelligence model. Then, the base station can obtain a channel estimation matrix at time t by inputting the channel estimation matrix at time t-1 and the received signal matrix at time t into the artificial intelligence model.

[0227] In one embodiment, the base station can calculate a channel estimation matrix based on an artificial intelligence model that receives received signal matrices for past SRS hierarchically and can drop the inputs by layer.

[0228] In one embodiment, a base station may selectively utilize past data based on auxiliary information related to channel quality, etc. For example, when utilizing an artificial intelligence model for channel estimation, the base station may use the auxiliary information to determine which past data is not helpful for channel estimation. Furthermore, the base station may not input some past data into the artificial intelligence model based on the determination result. The auxiliary information for the base station's determination may include information measured by a terminal and transmitted to the base station, and may include information directly measured by the base station.

[0229] In one embodiment, the base station may generate an indicator (index) corresponding to a channel estimation method based on auxiliary information, or receive an indicator corresponding to a channel estimation method from a terminal. The channel estimation method may indicate an artificial intelligence model for channel estimation, indicate whether historical data is utilized, or indicate a specific predefined channel estimation method (e.g., the number and type of historical data to be input, and the type of artificial intelligence model, etc.).

[0230] In one embodiment, the base station utilizes an outlier detection model to detect whether the received SRS is an outlier, and information related to the SRS detected as an outlier may not be used as input to an artificial intelligence model for channel estimation.

[0231] With respect to FIG. 17, the channel estimation of the base station can be performed based on the technical concept described in FIG. 4 to FIG. 16.

[0232] FIG. 18 illustrates an example of the operation of a terminal for estimating a channel according to embodiments of the present disclosure.

[0233] Referring to FIG. 18, the terminal can receive configuration information for an SRS from a base station in operation 1812. The configuration information may include information regarding frequency resources, time resources, scheduling information, or SRS transmission power for the terminal's SRS transmission.

[0234] In operation 1814, the terminal can transmit SRS based on configuration information received from the base station.

[0235] In operation 1816, the terminal can receive downlink data based on estimated channel information from the base station.

[0236] Regarding channel estimation of a base station, the terminal may transmit information directly measured or generated based on measured values ​​to the base station as auxiliary information for channel estimation. Alternatively, the terminal may generate an indicator to instruct the base station's channel estimation method based on the auxiliary information and transmit the generated indicator to the base station. Here, the channel estimation method may indicate an artificial intelligence model for channel estimation, indicate whether historical data is utilized for channel estimation, or indicate a specific predefined channel estimation method (e.g., the number and type of historical data to be input, the type of artificial intelligence model, the data input method, etc.).

[0237] According to the embodiments of the present disclosure, a base station can perform channel estimation with appropriate complexity and accuracy by utilizing an artificial intelligence model for channel estimation. Furthermore, the accuracy of channel estimation can be improved by utilizing various auxiliary information measurable or obtainable by the base station or terminal as input to the artificial intelligence model during the channel estimation process. In this case, regarding the utilization of auxiliary information, depending on the content of the auxiliary information, the auxiliary information itself may be used as input to the artificial intelligence model, or the auxiliary information may be used as reference information to determine the channel estimation method (type of artificial intelligence model, input, etc.). The embodiments of the present disclosure provide various procedures of the base station and terminal related to channel estimation and specific configurations of the artificial intelligence model for channel estimation.

[0238] Although the embodiments of the present disclosure have been described in detail to explain the technical concept of the present disclosure, the individual operations constituting each embodiment may be changed in order or parts of which may be omitted. Accordingly, an embodiment in which the order of operations is changed or some operations are omitted may be understood as having been described by the present disclosure. Furthermore, the embodiments of the present disclosure may be modified in various ways according to the content described in the present disclosure.

[0239] Although the operations of the method according to the embodiments of the present disclosure have been described separately for each embodiment, the operations included in each embodiment may be combined with the operations of other embodiments to form new embodiments. Therefore, embodiments in which the embodiments of the present disclosure are combined may also be understood as being described by the present disclosure.

[0240] The various embodiments of the present disclosure and the terms used therein are not intended to limit the technical features described in the present disclosure to specific embodiments, and should be understood to include various modifications, equivalents, or substitutions of said embodiments. In connection with the description of the drawings, similar reference numerals may be used for similar or related components. The singular form of a noun corresponding to an item may include one or more items unless the relevant context clearly indicates otherwise. In the present disclosure, each of the phrases such as “A or B”, “at least one of A and B”, “at least one of A or B”, “A, B or C”, “at least one of A, B and C”, and “at least one of A, B, or C” may include any one of the items listed together in the corresponding phrase, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used simply to distinguish a component from another component and do not limit the components in any other aspect (e.g., importance or order). Where a component (e.g., the first) is referred to as "coupled" or "connected" to another component (e.g., the second), with or without the terms "functionally" or "communicationally," it means that the component may be connected to the other component directly (e.g., via a wire), wirelessly, or through a third component.

[0241] As used in this disclosure, the term "module" may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be a component formed integrally, or a minimum unit of a component or part thereof that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0242] Various embodiments of the present disclosure may be implemented as software (e.g., a program) comprising one or more instructions stored in a storage medium (e.g., internal memory or external memory) readable by a machine (e.g., an electronic device). For example, a machine (e.g., a processor of an electronic device (e.g., processor (230)) may call at least one of the one or more instructions stored from the storage medium and execute it. This enables the machine to operate to perform at least one function according to at least one called instruction. One or more instructions may include code generated by a compiler or code that can be executed by an interpreter. The storage medium readable by the machine may be provided in the form of a non-transitory storage medium. Here, "non-transitory" simply means that the storage medium is a tangible device and does not contain a signal (e.g., an EM wave), and this term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily in the storage medium.

[0243] According to one embodiment, the method according to the various embodiments disclosed herein may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., CD-ROM (compact disc read-only memory)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created in a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0244] According to various embodiments, each component (e.g., module or program) of the described components may include a singular or multiple entities. According to various embodiments, one or more of the components or operations among the aforementioned components may be omitted, or one or more other components or operations may be added. Generally or additionally, multiple components (e.g., module or program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the multiple components in the same or similar manner as they were performed by the corresponding component among the multiple components prior to integration. According to various embodiments, operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically; one or more of the operations may be executed in a different order; omitted; or one or more other operations may be added.

Claims

1. A method performed by a base station in a wireless communication system, A step of transmitting configuration information for the SRS (sounding reference signal) to the terminal; A step of receiving a first SRS from the above terminal; A method comprising the step of estimating a channel for wireless communication using at least one artificial intelligence model based on a received first SRS, method.

2. In Claim 1, The step of estimating the channel for the above wireless communication is, The step of generating a first channel estimation matrix using a first artificial intelligence model based on the above first SRS, and A step comprising inputting a second channel estimated based on a second SRS received prior to the first SRS and the first channel into a second artificial intelligence model to output a third channel. method.

3. In Claim 1, The step of estimating the channel for the above wireless communication is, A step of generating a first feature vector using a third artificial intelligence model based on the first SRS above, and A step comprising inputting a second feature vector generated based on a second SRS received prior to the first SRS and the first feature vector into a fourth artificial intelligence model to generate a fourth channel estimation matrix, method.

4. In Claim 1, It further includes a step of obtaining auxiliary information for estimating the channel, and The above auxiliary information is, Including measurement information measured at the terminal or base station, information generated based on the measurement information, or an indicator relating to a method for estimating a channel, method.

5. In Claim 4, The above measurement information includes at least one of RSRP (reference signal received power), RSRQ (reference signal received quality), CQI (channel quality indicator), terminal speed, or Doppler shift. method.

6. In Claim 4, An indicator regarding the method for estimating the above channel is, A first indicator for indicating whether to estimate a channel based on a second SRS received prior to the first SRS, comprising method.

7. In Claim 6, The above-mentioned first indicator is, An outlier detection model trained based on a normal SRS receives the second SRS as input and is determined based on the output value, and The above measurement information includes a Doppler shift measured based on a slot including the second SRS, and The first indicator above is determined based on the comparison of the measured Dopler shift and an arbitrary threshold, method.

8. In Claim 5, An indicator regarding the method for estimating the above channel is, It includes a second indicator for indicating any one of a plurality of artificial intelligence models stored in the base station, and The second indicator above is determined based on the measurement information, method.

9. In a base station of a wireless communication system, At least one transceiver; At least one processor communicatively coupled to the above at least one transceiver; and It includes at least one memory that is communicationally coupled to the above at least one processor and stores instructions, and The above instructions are executed individually or in any combination by the above at least one processor, so that the base station: Transmit configuration information for the SRS (sounding reference signal) to the terminal, and Receive a first SRS from the above terminal, and Based on the received first SRS, at least one artificial intelligence model is used to estimate a channel for wireless communication, Base station.

10. In claim 9, the control unit is, Based on the above first SRS, a first channel estimation matrix is ​​generated using a first artificial intelligence model, and It is configured to generate a third channel estimation matrix by inputting the second channel estimation matrix, which is generated based on the second SRS received prior to the first SRS, and the first channel estimation matrix into the second artificial intelligence model, and Base station.

11. In claim 9, the control unit is, Based on the above-mentioned first SRS, a third artificial intelligence model is used to output a first feature vector, and A fourth channel estimation matrix is ​​generated by inputting a second feature vector generated based on a second SRS received prior to the first SRS and the first feature vector into a fourth artificial intelligence model. Base station.

12. In claim 9, the control unit is configured to acquire auxiliary information for estimating a channel, and The above auxiliary information is, Including measurement information measured at the terminal or base station, information generated based on the measurement information, or an indicator relating to a method for estimating a channel, Base station.

13. In Claim 12, The above measurement information includes at least one of RSRP (reference signal received power), RSRQ (reference signal received quality), CQI (channel quality indicator), terminal speed, or Doppler shift, and An indicator regarding the method for estimating the above channel is, It includes a second indicator for indicating any one of a plurality of artificial intelligence models stored in the base station, and The second indicator above is determined based on the measurement information, Base station.

14. In Claim 12, An indicator regarding the method for estimating the above channel is, A first indicator for indicating whether to estimate a channel based on a second SRS received prior to the first SRS, comprising Base station.

15. In Claim 14, The above-mentioned first indicator is, Determined based on the value output by an outlier detection model trained based on a normal SRS, which receives the second SRS as input. The above measurement information includes a Doppler shift measured based on a slot including the second SRS, and The first indicator above is determined based on the comparison of the measured Dopler shift and an arbitrary threshold, Base station.

Images