Channel estimation method, device, storage medium and program product

Abstract

This application provides a channel estimation method, apparatus, storage medium, and program product. The method includes: estimating the angle of a base station and a first scattering object surrounding the base station using a discrete Fourier transform algorithm to determine first angle information; designing a codebook-based transmit beamforming matrix based on the first angle information; and transmitting an uplink pilot signal to the base station based on the codebook-based transmit beamforming matrix. The solution of this application, based on the determined angle information, designs a codebook-based transmit beamforming matrix, improving the accuracy of uplink channel estimation and overcoming the limitations of channel estimation in traditional MIMO-OFDM systems.

Description

Technical Field

[0001] This application relates to the field of mobile communication technology, specifically to a channel estimation method, apparatus, storage medium, and program product. Background Technology

[0002] Orthogonal frequency division multiplexing (OFDM) effectively combats frequency-selective fading through subcarrier orthogonality, while multi-input multi-output (MIMO) technology utilizes spatial diversity to enhance system capacity. The combination of these two technologies, forming the MIMO-OFDM system, has become a core solution for integrated sensing systems. Existing research shows that sensing parameters such as angle and time delay are strongly correlated with channel state information (CSI), providing a theoretical basis for joint optimization.

[0003] The existing sensing-assisted uplink / downlink channel estimation schemes have the following limitations: (1) Insufficient joint sensing and channel estimation. Although existing methods based on sparse Bayesian learning or tensor decomposition can jointly estimate target parameters and channels, they do not fully exploit the spatiotemporal sparsity commonality of sensing channels and communication channels, resulting in insufficient utilization of sparsity. (2) Deficiencies in sensing-assisted channel estimation. Existing schemes (such as Kalman filtering and compressed sensing) are only designed for single time slots and do not utilize the quasi-static characteristics of channels in continuous time slots, resulting in a lack of time slot cooperation; uplink / downlink channel estimation is performed independently, lacking a cross-time slot parameter sharing mechanism, leading to redundant resource consumption.

[0004] In summary, existing channel estimation research for MIMO-OFDM integrated sensing systems is mostly limited to single time slots and does not fully utilize channel time correlation for joint uplink and downlink processing, resulting in limited improvement in channel estimation performance. Summary of the Invention

[0005] At least one embodiment of this application provides a channel estimation method, apparatus, storage medium, and program product to address the shortcomings of existing technologies.

[0006] To solve the above-mentioned technical problems, this application is implemented as follows:

[0007] In a first aspect, embodiments of this application provide a channel estimation method applied to a user equipment, comprising:

[0008] An angle estimate is performed on the base station and the first scattering object around the base station using the discrete Fourier transform algorithm to determine the first angle information;

[0009] Based on the first angle information, design a codebook-based transmit beamforming matrix;

[0010] Based on the codebook-based transmit beamforming matrix, an uplink pilot signal is transmitted to the base station.

[0011] Optionally, the method for estimating the angle of the base station and a first scattering object around the base station using a discrete Fourier transform algorithm to determine the first angle information includes:

[0012] Receive downlink reference signals sent by the base station;

[0013] Frequency domain information is obtained by performing a Discrete Fourier Transform (DFT) on the downlink reference signal using the DFT algorithm.

[0014] Based on the frequency domain information and the antenna array geometry, the angle corresponding to the maximum peak value is taken as the line-of-sight path angle of the base station, and the remaining peak values ​​represent the angle information of the first scatterer.

[0015] The line-of-sight path angle of the base station and the angle information of the first scatterer are determined as the first angle information.

[0016] Optionally, based on the first angle information, a codebook-based transmit beamforming matrix is ​​designed, including:

[0017] Based on the first angle information, a discretized set of angles covering the angle range is generated;

[0018] Based on the array antenna structure characteristics of the user equipment and the discretized angle set, a beamforming vector set corresponding to each discrete angle is generated to form a codebook-based transmit beamforming matrix.

[0019] Secondly, embodiments of this application provide a channel estimation method applied to a base station, including:

[0020] Receive uplink pilot signals sent by user equipment;

[0021] Perform a discrete Fourier transform on the uplink pilot signal to extract the second angle information of the user equipment and the second scattering body around the user equipment;

[0022] Based on the first angle information and the second angle information determined by the downlink, the orthogonal matching pursuit algorithm is dynamically optimized, and the channel estimation information is determined based on the optimized matching pursuit algorithm.

[0023] Optionally, a discrete Fourier transform is performed on the uplink pilot signal to extract the second angle information of the user equipment and the second scatterer around the user equipment, including:

[0024] The uplink pilot signal is subjected to discrete Fourier transform processing to obtain the frequency domain signal characteristics corresponding to the uplink pilot signal;

[0025] Based on the frequency domain signal characteristics, extract the set of peak components corresponding to the user equipment and the second scatterer around the user equipment;

[0026] Based on the phase difference characteristics of the signal received by the array antenna of the base station and the set of peak components, the estimated value of the angle of arrival corresponding to each peak component is calculated.

[0027] Based on the estimated arrival angle values ​​corresponding to each peak component, the main path angle of the user equipment and the reflection path angle of the second scatterer are combined to form the second angle information.

[0028] Optionally, based on the first angle information and the second angle information determined by the downlink, the orthogonal matching pursuit algorithm is dynamically optimized, and channel estimation information is determined based on the optimized matching pursuit algorithm, including:

[0029] The first angle information determined by the downlink and the second angle information extracted by the uplink are jointly analyzed to construct a fused angle feature set containing the main path of the user equipment and the reflection path of the second scatterer.

[0030] Based on the fused angle feature set, the search range and iteration step size of the orthogonal matching pursuit algorithm are dynamically adjusted, and atomic matching is performed within the effective angle range to determine the optimized matching pursuit algorithm;

[0031] Based on the optimized matching pursuit algorithm, the output includes channel impulse response estimation results containing multipath delay, angle, and complex gain.

[0032] Thirdly, embodiments of this application provide a channel estimation apparatus, applied to a user equipment, comprising:

[0033] The first determining module is used to perform angle estimation on the base station and the first scattering body around the base station using a discrete Fourier transform algorithm to determine the first angle information;

[0034] The first processing module is used to design a codebook-based transmission beamforming matrix based on the first angle information.

[0035] The second processing module is used to send uplink pilot signals to the base station based on the codebook-based transmit beamforming matrix.

[0036] Fourthly, embodiments of this application provide a channel estimation apparatus applied to a base station, comprising:

[0037] The receiving module is used to receive uplink pilot signals sent by user equipment;

[0038] The third processing module is used to perform a discrete Fourier transform on the uplink pilot signal to extract the second angle information of the user equipment and the second scatterer around the user equipment;

[0039] The fourth processing module is used to dynamically optimize the orthogonal matching pursuit algorithm based on the first angle information and the second angle information determined by the downlink, and to determine the channel estimation information based on the optimized matching pursuit algorithm.

[0040] Fifthly, embodiments of this application provide a computer-readable storage medium storing a program that, when executed by a processor, implements the steps of the method described in either the first or second aspect.

[0041] In a sixth aspect, embodiments of this application provide a computer program product, including computer instructions that, when executed by a processor, implement the steps of the method described in either the first or second aspect.

[0042] Compared with existing technologies, the channel estimation method, apparatus, storage medium, and program products provided in this application first utilize discrete Fourier transform to extract the scatterer angle information (first angle information) of the downlink on the base station side and construct a spatial feature library; then, based on this information, a codebook-based beamforming matrix is ​​dynamically designed to achieve directional transmission of the uplink pilot signal. This method of processing uplink and downlink angle information solves the existing problems through three key improvements: (1) extending single-slot estimation to cross-slot angle feature tracking, and establishing a channel time correlation model using the temporal stability of the scatterer's spatial position; (2) enhancing the effective coverage of the pilot signal and improving the signal-to-noise ratio through uplink and downlink joint beam optimization; and (3) reconstructing a sparse representation dictionary based on space-time dual constraints to reduce channel estimation errors. The solution in this application improves the channel tracking capability in high mobility scenarios. Attached Figure Description

[0043] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0044] Figure 1 This is a schematic diagram of a single-scenario synesthesia integration.

[0045] Figure 2 A flowchart illustrating a channel estimation method applied to user equipment is provided for embodiments of the present invention;

[0046] Figure 3 A flowchart illustrating a channel estimation method applied to a base station is provided for embodiments of the present invention;

[0047] Figure 4 This is a schematic diagram of the overall process of the channel estimation method provided in the embodiments of the present invention;

[0048] Figure 5 A comparative schematic diagram of normalized mean square error provided in an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of a channel estimation device applied to a user equipment is provided for an embodiment of the present invention;

[0050] Figure 7 A schematic diagram of the structure of a channel estimation device applied to a base station is provided for an embodiment of the present invention. Detailed Implementation

[0051] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, without limiting the number of objects; for example, the first object can be one or more. Furthermore, "or" in this application indicates at least one of the connected objects. For example, "A or B" covers three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0052] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc., in the instruction sent. An indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0053] As described in the background section, most existing studies focus only on channel estimation within a single uplink or downlink time slot, failing to utilize the cooperative potential between adjacent time slots. Due to the time correlation of channels, they typically exhibit quasi-static characteristics between consecutive uplink and downlink time slots. However, no research has yet explored sensing-assisted joint uplink and downlink processing to further enhance channel estimation performance. To address at least one of the above problems, embodiments of this application provide a channel estimation method, apparatus, storage medium, and program product that can reduce or avoid the occurrence of the above situations and improve channel estimation accuracy.

[0054] Reference Figure 1 As shown, an embodiment of this application provides a MIMO-OFDM sensing integrated system that operates in the millimeter-wave band and time-division duplex mode. The system model parameters and assumptions of this sensing integrated system are as follows: (1) In sensing operations, the user equipment is regarded as a point target; (2) Each base station is equipped with a uniform linear antenna array, with a transmit antenna and a receive antenna respectively; (3) Perfect synchronization is achieved between the base station and the user equipment through a global reference clock. The scenario setting is implemented here.

[0055] The system in this application also includes signal and channel models. In the case of signal transmission, in the system provided in this application, the base station and user equipment transmit OFDM signals to perform tasks, ensuring compatibility with existing wireless communication networks. Specifically, the transmitted analog time-domain signal is represented as follows:

[0056]

[0057] Where i = D or i = U corresponds to the transmission signal of the downlink or uplink, respectively; L i The index of a symbolic number can be represented as L. i ={0,…,L i -1},K i Δfi represents the number of subcarriers, and Δfi represents the subcarrier spacing. The duration of an OFDM symbol is... in, Indicates the duration of the basic symbol. Indicates the duration of the cyclic prefix. and These represent the frequency domain signals transmitted by the base station and the device on the k-th subcarrier and the l-th OFDM symbol, respectively. It is a rectangular function, when 0≤t≤T s When the condition is met, the function takes the value 1; otherwise, it takes the value 0.

[0058] In the downlink / uplink channels, due to the sparsity of millimeter-wave channels, this application employs a geometric channel model containing P paths to characterize the uplink and downlink channels. The downlink channel on the k-th subcarrier can be represented as:

[0059]

[0060] in, p = 0 represents the channel response of the line-of-sight path, p ∈ {1,…,P-1} corresponds to the non-line-of-sight path associated with the p-th scatterer, P represents the total number of multipaths, and its index set can be represented as P = {0,…,P-1}, ε p Let τ represent the equivalent path gain of the p-th path. p Let φ represent the delay of the p-th path. p Let θ represent the angle of arrival (AoA) of the p-th path. p Let AoD represent the departure angle (AoD) of the p-th path. Furthermore, This represents the downlink transmitter array response vector. This represents the downlink receiver array response vector.

[0061] Due to the channel heterogeneity of a time-division duplex system, the uplink channel is the transpose of the downlink channel, which can be expressed as:

[0062]

[0063] in,

[0064] In the case of signal reception, the downlink OFDM signal transmitted by the base station is received by the user equipment after propagation through the downlink channel, while the uplink OFDM signal transmitted by the user equipment is received by the base station after propagation through the uplink channel. Specifically, the frequency domain signal received by the base station or user equipment at the k-th subcarrier and the l-th OFDM symbol can be represented as:

[0065]

[0066] Where i = D or i = U represents the received signal of the downlink or uplink, respectively. Furthermore, This indicates the complex additive white Gaussian noise (AWGN) received by the device. This indicates the AWGN received by the BS.

[0067] Based on the above, refer to Figure 2 As shown, embodiments of this application provide a channel estimation method applied to a user equipment, including:

[0068] Step 21: The angle of the base station and the first scattering body around the base station is estimated by using the discrete Fourier transform algorithm to determine the first angle information.

[0069] In this embodiment, the channel estimation method applied to the user equipment is in the downlink phase. During the downlink phase, an angle estimation algorithm based on Discrete Fourier Transform (DFT) is used at the user equipment end to perform a coarse angle estimation of the base station and its surrounding scatterers (i.e., the first scatterer) to determine the first angle information. For example, the user equipment uses the DFT algorithm to analyze the multipath signals of the base station and surrounding scatterers, extracting the first angle information including AoA and AoD. This step captures time-frequency domain features simultaneously through two-dimensional transformation, overcoming the limitations of traditional single-slot estimation.

[0070] Step 22: Design a codebook-based transmit beamforming matrix based on the first angle information.

[0071] Here, a codebook-based transmit beamforming matrix is ​​dynamically constructed based on the extracted angular features, ensuring that the main lobe of the beam is precisely pointed towards the effective scattering path. This design significantly improves the signal-to-noise ratio of the pilot signal while reducing energy leakage in unnecessary directions.

[0072] Step 23: Based on the codebook-based transmit beamforming matrix, send uplink pilot signals to the base station.

[0073] In this embodiment, an optimized beamforming matrix is ​​used to transmit uplink pilot signals, ensuring that the base station receiver can accurately identify multipath component characteristics. This mechanism improves the efficiency of channel state information acquisition through dual spatial-temporal constraints. The method in this application improves channel estimation performance in complex scenarios by combining uplink and downlink angle information with dynamic beam control.

[0074] Optionally, step 21 above includes:

[0075] Receive downlink reference signals sent by the base station;

[0076] Frequency domain information is obtained by performing a Discrete Fourier Transform (DFT) on the downlink reference signal using the Discrete Fourier Transform (DFT) algorithm.

[0077] Based on the frequency domain information and antenna array geometry, the angle corresponding to the maximum peak value is taken as the base station line-of-sight path (LoS) angle, and the remaining peak values ​​represent the angle information of the first scatterer.

[0078] The line-of-sight path angle of the base station and the angle information of the first scatterer are determined as the first angle information.

[0079] In this application, the user equipment receives a known downlink reference signal transmitted by a base station, which contains spatial characteristic information of the base station antenna array. The time-domain reference signal is converted to a frequency-domain representation using DFT, and the spatial frequency components of the signal are obtained through complex spectrum analysis. Based on the antenna array geometry and the position of frequency-domain peaks, the angle corresponding to the maximum energy peak is resolved to the base station's LoS angle; other significant peaks correspond to the angles of arrival of multipath scatterers, reflecting the spatial scattering characteristics of the propagation environment. The LoS angle and the scatterer angles are then combined into the first angle information, providing a spatial channel feature library for subsequent beamforming. This mechanism significantly improves the angle resolution accuracy through joint time-frequency domain analysis.

[0080] Specifically, the downlink reference signal sent by the base station is received using the above formula (4), and the downlink reference signal is subjected to a DFT algorithm to obtain frequency domain information. Specifically, based on the received downlink reference signal, the downlink angle of arrival is estimated using the DFT method, i.e.,

[0081] Specifically, the frequency domain received signal of the user equipment on the υ-th antenna, the k-th subcarrier, and the l-th symbol is represented as follows:

[0082]

[0083] in, It can be seen It is a constant in the spatial dimension and is independent of the receiving antenna index υ.

[0084] The DFT-based angle estimation algorithm in the above steps can be expressed by the following formula:

[0085]

[0086] Where, n D ∈{0,…,N u -1}. Then, by applying peak detection, the estimated angle of arrival can be calculated as:

[0087]

[0088] in, This represents the peak index. Due to the dominant role of Loss of Speed ​​(LoS) in millimeter-wave channels, the angle corresponding to the maximum peak index is the angle of arrival relative to the base station, i.e.

[0089] Optionally, step 22 above includes:

[0090] Based on the first angle information, a discretized set of angles covering the angle range is generated;

[0091] Based on the array antenna structure characteristics of the user equipment and the discretized angle set, a beamforming vector set corresponding to each discrete angle is generated to form a codebook-based transmit beamforming matrix.

[0092] In this embodiment, step 22, as the core of the beamforming strategy design based on angle estimation, mainly includes two key sub-steps: angle set discretization and codebook-based beamforming matrix generation. These two steps together construct an optimized scheme suitable for directional transmission of uplink pilot signals. Based on the first angle information, the purpose of generating a discretized angle set covering the angle interval is to transform the continuous angle space into a finite number of discrete angle points for subsequent beamforming codebook construction. This is achieved by: based on the first angle information estimated in step 21 (such as the base station's AoA or scatterer angle distribution), determining an angle interval covering the main signal energy (e.g., a ±Δθ range centered on the estimated angle). This angle interval is then uniformly divided into multiple discrete angle points to form a discretized angle set. Here, discretization transforms the infinite-dimensional angle space into a finite-dimensional problem, reducing system complexity; the design of the coverage angle interval ensures effective coverage of the main signal path, even with estimation errors.

[0093] Furthermore, the purpose of forming the codebook-based transmit beamforming matrix is ​​to generate the optimal beamforming vector for each discrete angle, concentrating the uplink pilot signal energy in the target direction. Here, it is necessary to obtain the array antenna structural characteristics, such as the antenna array geometry (e.g., uniform linear array (ULA) or uniform planar array (UPA); antenna spacing d; and the number of antennas M, which can affect the beamforming gain. For each discrete angle, a corresponding beamforming vector is generated based on the array response function. All beamforming vectors are then arranged column-wise to form the codebook matrix.

[0094] The codebook matrix of this application predefines a set of optimal beam directions, from which the user equipment can directly select the beamforming vector that best matches the estimated angle; the beamforming vector designed by the array response function can maximize the signal gain in the target direction and suppress interference in other directions.

[0095] Specifically, based on the estimated angle information, a codebook-based transmit beamforming strategy is designed to achieve directional transmission of the uplink pilot signal. In this embodiment of the invention, considering the power limitations of user equipment and the significant propagation loss of millimeter-wave signals, the transmit beamforming is designed to point towards the base station to enhance uplink signal quality. However, due to the complexity of the channel environment and hardware limitations, angle estimation errors are inevitable. Therefore, the codebook-based transmit beamforming in the above steps can be expressed as:

[0096]

[0097] in, Let L represent the set of angles, and assume L... U >1, and assume that the angle estimation error for each path is limited to [-Δφ]. max ,Δφ max [This allows the beam direction to circumferentially rotate around the estimated angle.] Dynamic adjustments are made to improve beam alignment accuracy and enhance uplink robustness.

[0098] Existing technologies may use fixed codebooks (such as DFT codebooks), which cannot adaptively adjust the angle coverage range. This step dynamically generates the coverage area based on real-time estimated angle information, improving adaptability to dynamic channels. Existing solutions may ignore the actual array structure (such as antenna spacing and arrangement), leading to a decrease in beamforming performance. This step explicitly considers array characteristics, and the generated codebook is more closely aligned with the actual hardware configuration. By covering angle ranges rather than single-point angles, the codebook can tolerate a certain degree of angle estimation error (such as the coarse estimation error in step 21), ensuring robustness in practical applications.

[0099] The uplink pilot signal transmitted in this application enhances the signal strength at the base station receiver, improves channel estimation accuracy, and thus improves communication quality.

[0100] Reference Figure 3 As shown, this application provides a channel estimation method applied to a base station, including:

[0101] Step 31: Receive the uplink pilot signal sent by the user equipment.

[0102] Here, step 31 involves acquiring a reference signal for channel estimation, providing a data foundation for subsequent angle extraction and parameter optimization. For example, the pilot signal design can employ orthogonal or pseudo-random sequences to ensure good autocorrelation characteristics in the time-frequency domain; the pilot density must satisfy the Nyquist sampling theorem. Correspondingly, the multi-antenna array on the base station side can synchronously receive signals through step 31, forming a received signal matrix. Existing solutions may rely on fixed pilot patterns, while this step supports dynamically adjusting the pilot position based on downlink feedback, such as increasing pilot density in high Doppler regions.

[0103] Step 32: Perform Discrete Fourier Transform (DFT) on the uplink pilot signal to extract the second angle information of the user equipment and the second scatterer around the user equipment.

[0104] In this embodiment, performing a Directed Fourier Transform (DFT) on the uplink pilot signal allows for an N-point DFT on the received signal of each antenna, converting it to the frequency domain to obtain the frequency domain signal characteristics corresponding to the uplink pilot signal. Peak detection is then performed on these frequency domain signal characteristics, and the frequency index corresponding to the maximum peak is mapped to an angle. If multiple peaks exist, corresponding to multipath signals, each peak corresponds to a scatterer angle. Using the frequency index corresponding to the maximum peak mapped to an angle and the multiple scatterer angles, the second angle information of the user equipment and the second scatterers surrounding the user equipment is determined.

[0105] Step 33: Based on the first angle information and the second angle information determined by the downlink, dynamically optimize the orthogonal matching pursuit algorithm, and determine the channel estimation information based on the optimized matching pursuit algorithm.

[0106] It should be noted that the sensing-aided orthogonal matching pursuit (SAOMP) algorithm is a sparse signal reconstruction algorithm that iteratively selects the atoms most relevant to the residual (corresponding to channel multipath) to gradually approximate the channel impulse response.

[0107] In this embodiment, the first angle information is the downlink estimated angle of scatterers around the base station; the second angle information is the uplink estimated angle of scatterers around the user equipment. The first and second angle information together constitute a multipath angle pair. Using the multipath angle pair allows only the array response atoms corresponding to the angle pair to be retained, reducing the atom space and redundant searching. The angle estimation error is determined using the multipath angle pair, such as the downlink and uplink angle errors. A weight factor is added to each atom; the smaller the error, the higher the weight, and high-weight atoms are preferentially selected. Utilizing the angle estimation error and the sparsity of the angle information, the orthogonal matching pursuit algorithm is dynamically optimized. For example, in millimeter-wave channels, the multipath number P << M, and the preset sparsity K = P + Q, terminating after K iterations to avoid overfitting. The optimized SAOMP algorithm outputs the index set of non-zero atoms and the corresponding channel gain, ultimately outputting the channel matrix.

[0108] Specifically, the parameter configuration example for the dynamically optimized orthogonal matching pursuit algorithm is as follows: number of DFT points (N=256), number of antennas (M=64), angle estimation error threshold of ±5°, and upper limit of SAOMP iteration count (K=16) (corresponding to the number of multipaths (P=8) and the number of downlink angles (Q=8)).

[0109] This application achieves a complementary relationship between the downlink first angle (base station view) and the uplink second angle (user equipment view), reducing the multipath search space and improving the convergence speed of SAOMP. DFT peak detection combined with angle prior information reduces noise interference in angle estimation. Angle constraints reduce the atomic space dimension, making it suitable for real-time channel estimation in large-scale MIMO systems (such as 128-antenna base stations). For example, this application can be applied to millimeter-wave communication systems (28GHz / 60GHz) to improve channel estimation efficiency by utilizing angle sparsity; it can also be applied to mobile scenarios (such as high-speed rail and drone communication) to adapt to rapidly changing channels through dynamic angle updates.

[0110] Optionally, step 32 above includes:

[0111] The uplink pilot signal is subjected to discrete Fourier transform processing to obtain the frequency domain signal characteristics corresponding to the uplink pilot signal;

[0112] Based on the frequency domain signal characteristics, extract the set of peak components corresponding to the user equipment and the second scatterer around the user equipment;

[0113] Based on the phase difference characteristics of the signal received by the array antenna of the base station and the set of peak components, the estimated value of the angle of arrival corresponding to each peak component is calculated.

[0114] Based on the estimated arrival angle values ​​corresponding to each peak component, the main path angle of the user equipment and the reflection path angle of the second scatterer are combined to form the second angle information.

[0115] In this embodiment, an N-point DFT transformation is performed on the uplink pilot signals received by each antenna of the base station to the frequency domain, obtaining the frequency domain signal characteristics corresponding to the uplink pilot signals. Here, the time-domain signal is converted into a frequency-domain representation to separate different frequency components, facilitating subsequent peak detection. Frequency domain signal characteristics, such as amplitude and phase, can directly reflect the frequency-selective fading and multipath effects of the channel. Based on the frequency domain signal characteristics, the amplitude of each frequency point is calculated, and a threshold is set, such as the mean plus 3 times the standard deviation, to filter out frequency points with amplitudes exceeding the threshold. Non-maximum suppression is performed on adjacent peaks to ensure that each multipath corresponds to a unique peak point, and each peak component corresponds to an independent signal path (line-of-sight path or scattering path), ultimately outputting a set of peak components. The arrival angle is calculated based on the array phase difference and the peak components. The angle corresponding to the largest peak amplitude is taken as the main path angle of the user equipment, and the angles corresponding to the remaining peaks are the reflection path angles of the second scatterer. The second angle information is constructed by merging the main path angle and the reflection path angle of the scatterer into a set.

[0116] The main path angle in the second angle information of this application is used for beamforming main direction alignment; the scattering path angle is used for multipath aggregation or channel modeling.

[0117] Specifically, in the uplink phase, the base station estimates the angle information of the user equipment and its surrounding scattering objects based on the received pilot signal using the DFT method; in this embodiment, the received frequency domain signal in the above formula (4) is used to estimate the uplink angle of arrival using DFT technology, i.e. Specifically, the base station in the The frequency domain received signal on the k-th subcarrier and the l-th symbol is:

[0118]

[0119] in, It is worth noting that this is related to the omnidirectional transmission signal sent by the base station. Unlike other systems, user equipment uses transmit beamforming to send uplink pilot signals. Therefore, It can be further expressed as It can be seen that, With receiving antenna index Irrelevant. Therefore, the received signal on all antennas... Apply N in the above steps b Point DFT, and the application of peak probing, can effectively estimate the uplink angle of arrival, which can be expressed as: Due to the dominant influence of the line-of-sight path, the angle corresponding to the maximum peak index is the estimated angle of arrival relative to the device, i.e.,

[0120] Optionally, step 33 above includes:

[0121] The first angle information determined by the downlink and the second angle information extracted by the uplink are jointly analyzed to construct a fused angle feature set containing the main path of the user equipment and the reflection path of the second scatterer.

[0122] Based on the fused angle feature set, the search range and iteration step size of the orthogonal matching pursuit algorithm are dynamically adjusted, and atomic matching is performed within the effective angle range to determine the optimized matching pursuit algorithm;

[0123] Based on the optimized matching pursuit algorithm, the output includes channel impulse response estimation results containing multipath delay, angle, and complex gain.

[0124] In this embodiment, the first angle information and the second angle information are jointly analyzed to construct a fused angle feature set, integrating the angle information of the uplink and downlink to form a more complete description of the channel multipath characteristics, providing accurate angle priors for subsequent SAOMP algorithm optimization. The first angle information is the angle of the base station and surrounding scatterers estimated by the user equipment in the downlink (e.g., downlink AoA); the second angle information is the angle of the user equipment and surrounding scatterers estimated by the base station in the uplink (e.g., uplink AoA). Through spatial geometric relationships, the uplink and downlink angles are associated as "AoA-AoD" pairs of multipaths, for example, the AoA and AoD of the line-of-sight path satisfy a mirror symmetry relationship. Outliers with angle estimation errors exceeding a threshold (e.g., ±5°) are removed, and valid angle pairs are selected. The angles of the scatterer reflection paths are clustered to form a fused set of LoS and K scattering paths, merging overlapping angle intervals. Here, the joint analysis of uplink and downlink angles can reduce the angle estimation error and avoid the one-sidedness of single-link angle estimation; the fused feature set directly reflects the multipath topology of the channel, providing accurate atomic spatial constraints for the SAOMP algorithm. Based on the fusion angle set, the atomic space of the SAOMP algorithm is limited to the effective angle range, excluding array response atoms corresponding to invalid angles. The atomic space dimension is reduced from (O(MN)) to (O(K)) (where M is the number of antennas and N is the number of frequency points), reducing the computational complexity by a factor of MN / K (K is the number of multipaths). For main path angle pairs, a smaller iteration step size (e.g., 0.1°) is set to improve estimation accuracy. For scattering path angle pairs, the step size is dynamically adjusted according to the angle estimation error (the larger the error, the larger the step size, accelerating convergence). Here, a preset iteration step size formula can be used for iterative calculation. The number of iterations is reduced from 20 in standard SAOMP to 8-12, improving the convergence speed. The multipath resolution is improved, enabling the differentiation of dense multipath paths with angular intervals less than 1°. Multipath delay is calculated using frequency domain peak indexing and sampling frequency; the fused angle pairs correspond to the incoming and outgoing directions of the multipath, determining the angle information; the channel amplitude and phase are estimated using the least squares method, determining the complex gain. A preset impulse function is used to construct the channel impulse response (CIR). The estimation results determined by the above steps can be directly used for beamforming in millimeter-wave communication (e.g., the main path angle is used for beam orientation); the multipath delay and complex gain provide accurate parameters for channel modeling (e.g., the Saleh-Valenzuela model), supporting time-varying channel prediction.

[0125] Specifically, based on the estimated angle information, a codebook-based receiver combiner is designed to improve the received signal quality. In this embodiment, to improve the received signal quality, the base station designs a receiver combiner based on the estimated AoD to enhance the desired signal. Considering the imperfections of angle estimation, it is assumed that the angle estimation error for each path is limited to [-Δθ]. max ,Δθ maxThe codebook of the receiver combiner in the above steps can be constructed as follows: in, Θ represents the set of angles, which can be expanded as

[0126] Based on angle information, the sparse representation matching strategy in the orthogonal matched pursuit process is dynamically optimized using sensing information to achieve high-precision channel state information estimation. This includes a sensing-assisted sparse signal reconstruction method. To improve the accuracy of channel estimation, the device sequentially transmits pilot signals. The base station performs channel response measurements under different transmit and receive parameters configured based on the angle information obtained from sensing. Specifically, the received signal of the base station on the k-th subcarrier and the l-th symbol can be expressed as:

[0127]

[0128] in, Equation (a) is based on This is derived. Then, stack L. U The received signals over consecutive symbols yield:

[0129]

[0130] in,

[0131] The channel matrix in formula (3) above It can be compactly represented in matrix form as follows:

[0132]

[0133] in, and This represents the array response matrix, where each column represents the array response vector. Furthermore, in, The channel gain is represented by the angle. Due to the dominant role of Loss of Light (LoS) and the limited number of strong scatterers in millimeter-wave propagation, the channel exhibits significant sparsity in the angular domain. This sparsity can be characterized by discretizing the angular domain into a finite set of grid points, where each grid point represents a possible AoD or AoA. Using the estimated angle as prior information, this invention dynamically constructs the grid to focus on the main propagation path. Specifically, based on the estimated angle, the grids corresponding to the uplink AoD (i.e., downlink AoA) and uplink AoA are defined as follows:

[0134]

[0135] Where Q = {0, ..., Q-1}, Q > 1 represents the number of grid points. Both grids Ω and Γ contain G = PQ >> P grid points. Utilizing sparsity, the channel matrix... It can be represented in the form of a sparse angular domain:

[0136]

[0137] in, and This represents the array response matrix corresponding to the grid. and These represent the grid points corresponding to the uplink arrival angle and departure angle, respectively. Let represent a sparse matrix containing P non-zero elements, representing the channel gain. Then, using the channel matrix in (14), the received signal... It can be transformed into:

[0138]

[0139] in, Let represent a sparse vector containing P non-zero elements. Represents the measurement matrix. Given Due to sparsity, channel estimation can be transformed into a sparse signal recovery problem, which can be solved using compressed sensing-based methods.

[0140] Finally, this application provides a detailed explanation of the sensing-aided orthogonal matching pursuit (SAOMP) algorithm, which optimizes the orthogonal matching pursuit process by estimating the angle, thereby achieving high-precision channel state information estimation.

[0141] In the SAOMP algorithm, the input information is: the received signal. Angle estimate and Residual Iteration Channel gain Index Set The iteration index j = 0. The output information is:

[0142] The first step of the SAOMP algorithm is to generate a mesh based on the sensing results.

[0143] Based on the estimated angle, a grid corresponding to the uplink AoD is generated. and the grid corresponding to the uplink AoA

[0144] Generate array response matrix based on grid. and

[0145] Generate measurement matrix in,

[0146] Step 2 of the SAOMP algorithm is to obtain the non-zero channel gain value. When j ≤ P-1, j = j+1; Stop the loop.

[0147] Step 3 of the SAOMP algorithm is to obtain the channel gain vector. When g = 1, ..., G 2 If g∈J, otherwise, Stop the loop.

[0148] Step 4 of the SAOMP algorithm is: generating the channel matrix.

[0149] In step 1, the angle information obtained by sensing is used to construct grids Ω and Γ corresponding to the uplink AoD and AoA, and a measurement matrix is ​​generated accordingly. In each iteration of step 2, step b) selects the measurement matrix. Neutral and residual The column with the strongest correlation. In step c), update the column index set J. (j) =, where each element in this set corresponds to an angle of arrival / pair in the grid. Then, in step d), the channel gain corresponding to the selected angle of arrival / departure pair is estimated using a least-squares method. In step e), the residuals are updated by subtracting the contribution of the selected column. This iterative process continues until a preset sparsity level is reached, at which point all non-zero channel gains are determined. Next, in step 3, the channel gain vector is obtained based on the non-zero channel gain values. Finally, in step 4, a high-precision channel estimate is constructed based on the array response matrix and the estimated channel gain vector, denoted as .

[0150] Traditional compressed sensing-based channel estimation methods rely on a fixed grid obtained by uniformly quantizing the angle domain [0, π). In contrast, this invention dynamically constructs the grid based on the estimated angles, thereby focusing on the primary propagation paths. Specifically, by limiting the angle domain to the ranges of the departure and arrival angles, respectively... and This invention significantly reduces the grid size while increasing the grid resolution, thereby effectively reducing computational complexity and improving estimation accuracy.

[0151] Reference Figure 4The overall flowchart shown can coordinate the uplink and downlink. This application embodiment designs a frame structure and signal processing flowchart following a time-division duplex mechanism, as shown below. Figure 4 As shown, each frame is divided into three phases: Phase I for downlink pilot signal transmission, Phase II for uplink pilot signal transmission, and Phase III for uplink or downlink data transmission. This invention focuses on the first two phases, aiming to achieve high-precision uplink channel estimation. Specifically, in Phase I, due to a lack of prior information about the device, the base station transmits omnidirectional pilot signals, while the device receives downlink signals via the Loss of Position (LoS) and non-line-of-sight (NLoS) paths. Considering computational resource limitations, the device only estimates the AoD (Aspect-Oriented Distance) to obtain coarse angle information. Using these preliminary estimates, and combining them with the temporal correlation of the channel, the device designs transmit beamforming to enhance uplink signal quality. In Phase II, the base station estimates the AoD based on the received pilot signals and designs a combiner accordingly. Subsequently, the base station performs refined channel estimation using the estimated angular domain information to obtain high-precision channel state information.

[0152] Furthermore, Figure 5 This is a comparative diagram illustrating the normalized mean square error (NMSE) provided in an embodiment of the present invention. (See reference...) Figure 5 As shown, the performance of the proposed SAOMP algorithm is compared with that of the least-squares (LS) method, the minimum mean square error (MMSE) method, and the beam training-based MMSE method. The horizontal axis represents the signal-to-noise ratio (SNR) in decibels, and the vertical axis represents the NMSE. It can be seen that the NMSE gradually decreases as the SNR increases. Furthermore, the proposed SAOMP algorithm exhibits superior estimation performance under all SNR conditions. Even in low SNR environments, the algorithm achieves the lowest NMSE, demonstrating strong robustness in harsh scenarios. Simultaneously, the estimation accuracy of the proposed algorithm also significantly improves with increasing SNR, further expanding its performance advantage over other schemes.

[0153] In summary, this application proposes an uplink channel estimation method based on perception-assisted and downlink cooperation. It fully utilizes the temporal correlation between adjacent time slots and designs a two-stage channel estimation method to obtain high-precision uplink channel state information. First, in the downlink stage, an angle estimation algorithm based on Discrete Fourier Transform (DFT) is used to perform a coarse angle estimation of the base station and surrounding scatterers at the device end. Based on this, a codebook-based transmit beamforming is designed to directionally transmit the uplink pilot signal to the base station. Subsequently, in the uplink stage, the base station uses the DFT algorithm to obtain the angles of the device and scatterers, and based on this, a codebook-based receiver combiner is designed to enhance signal reception quality. Finally, this application also proposes a perception-assisted Orthogonal Matching Pursuit (SAOMP) algorithm, which optimizes the orthogonal matching pursuit process using the estimated angle, thereby achieving high-precision channel state information estimation.

[0154] The various methods described above are based on embodiments of this application. Apparatus for implementing the above methods will now be provided.

[0155] Please refer to Figure 6 This application also provides a channel estimation apparatus, applied to a user equipment, comprising:

[0156] The first determining module 61 is used to perform angle estimation on the base station and the first scattering body around the base station using a discrete Fourier transform algorithm to determine the first angle information;

[0157] The first processing module 62 is used to design a codebook-based transmission beamforming matrix based on the first angle information.

[0158] The second processing module 63 is used to send uplink pilot signals to the base station based on the codebook-based transmit beamforming matrix.

[0159] Optionally, the first determining module 61 described above includes:

[0160] The receiving unit is used to receive downlink reference signals sent by the base station;

[0161] The unit is used to perform a discrete Fourier transform on the downlink reference signal using a discrete Fourier transform algorithm to obtain frequency domain information;

[0162] The first processing unit is used to take the angle corresponding to the maximum peak value as the line-of-sight path angle of the base station based on the frequency domain information and the antenna array geometry, and the remaining peak values ​​represent the angle information of the first scatterer.

[0163] The first determining unit is used to determine the line-of-sight path angle of the base station and the angle information of the first scatterer as the first angle information.

[0164] Optionally, the first processing module 62 described above includes:

[0165] The second processing unit is used to generate a discretized set of angles covering the angle range based on the first angle information.

[0166] The third processing unit is used to generate a set of beamforming vectors corresponding to each discrete angle based on the array antenna structure characteristics of the user equipment and the set of discrete angles, forming a codebook-based transmit beamforming matrix.

[0167] It should be noted that the device in this embodiment corresponds to the method applied to the user device side described above. The implementation methods in each of the above embodiments are also applicable to the embodiments of this device and can achieve the same technical effect. The device provided in this application embodiment can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Therefore, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail here.

[0168] Please refer to Figure 7 This application also provides a channel estimation apparatus applied to a base station, comprising:

[0169] The receiving module 71 is used to receive the uplink pilot signal sent by the user equipment;

[0170] The third processing module 72 is used to perform a discrete Fourier transform on the uplink pilot signal to extract the second angle information of the user equipment and the second scattering body around the user equipment;

[0171] The fourth processing module 73 is used to dynamically optimize the orthogonal matching pursuit algorithm based on the first angle information and the second angle information determined by the downlink, and to determine the channel estimation information based on the optimized matching pursuit algorithm.

[0172] Optionally, the third processing module 72 described above includes:

[0173] The acquisition unit is used to perform discrete Fourier transform processing on the uplink pilot signal to obtain the frequency domain signal characteristics corresponding to the uplink pilot signal.

[0174] The extraction unit is used to extract the set of peak components corresponding to the user equipment and the second scatterer around the user equipment based on the frequency domain signal characteristics;

[0175] The calculation unit is used to calculate the estimated angle of arrival for each peak component based on the phase difference characteristics of the signal received by the array antenna of the base station and the set of peak components.

[0176] The fourth processing unit is used to combine the main path angle of the user equipment and the reflection path angle of the second scatterer to form the second angle information based on the estimated arrival angle values ​​corresponding to each peak component.

[0177] Optionally, the fourth processing module 73 described above includes:

[0178] The fifth processing unit is used to jointly analyze the first angle information determined by the downlink and the second angle information extracted by the uplink to construct a fused angle feature set containing the main path of the user equipment and the reflection path of the second scatterer.

[0179] The sixth processing unit is used to dynamically adjust the search range and iteration step size of the orthogonal matching pursuit algorithm based on the fused angle feature set, and to perform atomic matching within the effective angle range to determine the optimized matching pursuit algorithm;

[0180] The seventh processing unit is used to output channel impulse response estimation results, including multipath delay, angle, and complex gain, based on the optimized matching pursuit algorithm.

[0181] It should be noted that the device in this embodiment corresponds to the method applied to the base station side described above. The implementation methods in each of the above embodiments are also applicable to the embodiments of this device and can achieve the same technical effect. The device provided in this application embodiment can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0182] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described functionality. Figure 2 or Figure 3 The various processes of the illustrated method embodiments achieve the same technical effect, and will not be described again here to avoid repetition. The computer-readable storage medium mentioned includes, for example, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0183] This application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the above-described... Figure 2 or Figure 3 The various processes of the method embodiments shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0184] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0185] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0186] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A channel estimation method, characterized in that, Applied to user equipment, including: An angle estimate is performed on the base station and the first scattering object around the base station using the discrete Fourier transform algorithm to determine the first angle information; Based on the first angle information, design a codebook-based transmit beamforming matrix; Based on the codebook-based transmit beamforming matrix, an uplink pilot signal is transmitted to the base station.

2. The method according to claim 1, characterized in that, This is used to estimate the angle of the base station and a first scattering object around the base station using a discrete Fourier transform algorithm, and to determine the first angle information, including: Receive downlink reference signals sent by the base station; Frequency domain information is obtained by performing a Discrete Fourier Transform (DFT) on the downlink reference signal using the DFT algorithm. Based on the frequency domain information and the antenna array geometry, the angle corresponding to the maximum peak value is taken as the line-of-sight path angle of the base station, and the remaining peak values ​​represent the angle information of the first scatterer. The line-of-sight path angle of the base station and the angle information of the first scatterer are determined as the first angle information.

3. The method according to claim 1, characterized in that, Based on the first angle information, a codebook-based transmit beamforming matrix is ​​designed, including: Based on the first angle information, a discretized set of angles covering the angle range is generated; Based on the array antenna structure characteristics of the user equipment and the discretized angle set, a beamforming vector set corresponding to each discrete angle is generated to form a codebook-based transmit beamforming matrix.

4. A channel estimation method, characterized in that, Applied to base stations, including: Receive uplink pilot signals sent by user equipment; Perform a discrete Fourier transform on the uplink pilot signal to extract the second angle information of the user equipment and the second scattering body around the user equipment; Based on the first angle information and the second angle information determined by the downlink, the orthogonal matching pursuit algorithm is dynamically optimized, and the channel estimation information is determined based on the optimized matching pursuit algorithm.

5. The method according to claim 4, characterized in that, Perform a Discrete Fourier Transform on the uplink pilot signal to extract the second angle information of the user equipment and the second scatterer around the user equipment, including: The uplink pilot signal is subjected to discrete Fourier transform processing to obtain the frequency domain signal characteristics corresponding to the uplink pilot signal; Based on the frequency domain signal characteristics, extract the set of peak components corresponding to the user equipment and the second scatterer around the user equipment; Based on the phase difference characteristics of the signal received by the array antenna of the base station and the set of peak components, the estimated value of the angle of arrival corresponding to each peak component is calculated. Based on the estimated arrival angle values ​​corresponding to each peak component, the main path angle of the user equipment and the reflection path angle of the second scatterer are combined to form the second angle information.

6. The method according to claim 4, characterized in that, Based on the first angle information and the second angle information determined by the downlink, the orthogonal matching pursuit algorithm is dynamically optimized. Channel estimation information is then determined based on the optimized matching pursuit algorithm, including: The first angle information determined by the downlink and the second angle information extracted by the uplink are jointly analyzed to construct a fused angle feature set containing the main path of the user equipment and the reflection path of the second scatterer. Based on the fused angle feature set, the search range and iteration step size of the orthogonal matching pursuit algorithm are dynamically adjusted, and atomic matching is performed within the effective angle range to determine the optimized matching pursuit algorithm; Based on the optimized matching pursuit algorithm, the output includes channel impulse response estimation results containing multipath delay, angle, and complex gain.

7. A channel estimation device, characterized in that, Applied to user equipment, including: The first determining module is used to perform angle estimation on the base station and the first scattering body around the base station using a discrete Fourier transform algorithm to determine the first angle information; The first processing module is used to design a codebook-based transmission beamforming matrix based on the first angle information. The second processing module is used to send uplink pilot signals to the base station based on the codebook-based transmit beamforming matrix.

8. A channel estimation device, characterized in that, Applied to base stations, including: The receiving module is used to receive uplink pilot signals sent by user equipment; The third processing module is used to perform a discrete Fourier transform on the uplink pilot signal to extract the second angle information of the user equipment and the second scatterer around the user equipment; The fourth processing module is used to dynamically optimize the orthogonal matching pursuit algorithm based on the first angle information and the second angle information determined by the downlink, and to determine the channel estimation information based on the optimized matching pursuit algorithm.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 3, or implements the steps of the method as described in any one of claims 4 to 6.

10. A computer program product, characterized in that, Includes computer instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1 to 3, or implement the steps of the method as claimed in any one of claims 4 to 6.

Images