Channel estimation method and device for wireless communication, electronic equipment and storage medium

Abstract

The invention relates to the field of wireless communication, in particular to a channel estimation method and device for wireless communication, electronic equipment and a storage medium, and the method comprises the steps: extracting a first channel characteristic matrix in each scale and each direction from an initial channel state matrix contained in pilot frequency information; inputting the latest first channel characteristic matrix into a denoising network to obtain a second channel characteristic matrix; correcting the second channel characteristic matrix according to a noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain a latest first channel characteristic matrix; denoising and correction are continued until the preset denoising frequency is reached; and determining a target channel state matrix corresponding to the pilot frequency information based on all the finally obtained first channel characteristic matrixes. According to the channel estimation method and device, channel estimation is carried out in the mode that noise is removed and corrected through multiple times of iteration, the channel estimation performance can be guaranteed under the condition that the noise level is high or the channel condition changes greatly, and robustness is improved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communications, and in particular to a channel estimation method, device, electronic device, and storage medium for wireless communications. Background Art

[0002] In wireless communication systems, channel estimation is a key technology for ensuring communication quality. Essentially, it involves accurately modeling the transmission characteristics of wireless channels and estimating their parameters using signal processing techniques. Wireless channels are susceptible to factors such as multipath fading, the Doppler effect, and noise interference. This makes it crucial for the receiver to accurately estimate channel state information to ensure effective signal demodulation and recovery.

[0003] Traditional channel estimation methods often use end-to-end neural networks, whose input is pilot information and output is channel estimation results.

[0004] However, when the noise level is high or the channel conditions vary greatly, the evaluation performance of traditional methods will degrade significantly and their robustness will be insufficient. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a channel estimation method, device, electronic device and storage medium for wireless communication, which can ensure channel estimation performance and improve robustness when the noise level is high or the channel conditions vary greatly.

[0006] In a first aspect, an embodiment of the present application provides a channel estimation method for wireless communication, the method comprising: Extracting a first channel characteristic matrix at each scale and in each direction from an initial channel state matrix included in the pilot information; For each first channel characteristic matrix, inputting the latest first channel characteristic matrix into a denoising network to obtain a second channel characteristic matrix; the denoising network is trained based on an actual channel state matrix and a complete noisy channel characteristic matrix corresponding to the actual channel state matrix obtained using a noise scheduling parameter; the noise scheduling parameter is used to control the noise addition ratio; Correcting the second channel characteristic matrix according to a noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain an updated first channel characteristic matrix; and continuously inputting the updated first channel characteristic matrix into the denoising network until a preset number of denoising cycles is reached; Based on all the first channel characteristic matrices finally obtained, a target channel state matrix corresponding to the pilot information is determined.

[0007] In a possible implementation, the correcting the second channel characteristic matrix according to the noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain the latest first channel characteristic matrix includes: determining a noise error adjustment strength according to a numerical relationship between the noise error value and a noise error threshold; The second channel characteristic matrix is corrected according to the noise error adjustment strength and the noise error value to obtain the latest first channel characteristic matrix.

[0008] In a possible implementation, determining the noise error adjustment strength according to a numerical relationship between the noise error value and a noise error threshold includes: Determining whether the noise error value is greater than the noise error threshold; If the noise error value is greater than the noise error threshold, calculating a noise error adjustment strength according to the noise error value; If the noise error value is less than or equal to the noise error threshold, the noise error adjustment strength is set to 1.

[0009] In a possible implementation, calculating the noise error adjustment strength according to the noise error value includes: Substituting the noise error value into the following formula, the noise error adjustment strength is obtained; ; in, Adjust the strength for noise error, is the noise error value, is the remaining number of processing times.

[0010] In a possible implementation, the modifying the second channel characteristic matrix according to the noise error adjustment strength and the noise error value to obtain the latest first channel characteristic matrix includes: Substituting the noise error adjustment strength and the noise error value into the following formula to obtain the latest first channel characteristic matrix; ; in, is the latest first channel feature matrix, is the second channel feature matrix, is the remaining number of processing times, is the denoising network, are the trainable parameters of the denoising network, is the latest first channel feature matrix obtained when the remaining processing times is t+1.

[0011] In one possible implementation, the denoising network is trained by the following steps: Extracting the actual channel feature matrix of the channel state information at each scale and in each direction from the actual channel state matrix; Iteratively adding noise to the actual channel characteristic matrix according to the noise scheduling parameters to obtain a complete noise channel characteristic matrix; After denoising and correcting the complete noisy channel characteristic matrix through the denoising network, a restored channel state matrix corresponding to the complete noisy channel characteristic matrix is obtained; The parameters in the denoising network are updated with the minimum difference between the restored channel state matrix and the actual channel state matrix as the optimization goal; and the completely noisy channel characteristic matrix is continuously denoised and corrected through the denoising network until a preset iteration stop condition is met.

[0012] In a possible implementation, the iteratively adding noise to the actual channel characteristic matrix according to the noise scheduling parameter to obtain a complete noisy channel characteristic matrix includes: Inputting the noise scheduling parameters and the actual channel characteristic matrix into the following formula to obtain a complete noise channel characteristic matrix; ; ; in, is the complete noise channel characteristic matrix, is the noise scheduling parameter of the i-th iteration, The number of times noise is added for a preset iteration, is the noise scheduling parameter for the j+1th iteration, is the standard complex Gaussian noise value of the jth iteration that follows the standard normal distribution , I is the covariance matrix of the standard normal distribution, is the actual channel characteristic matrix.

[0013] In a second aspect, an embodiment of the present application further provides a channel estimation device for wireless communication, the device comprising: An extraction module, configured to extract a first channel characteristic matrix at each scale and in each direction from an initial channel state matrix included in the pilot information; An input module is configured to input the latest first channel characteristic matrix into a denoising network for each first channel characteristic matrix to obtain a second channel characteristic matrix; the denoising network is trained based on an actual channel state matrix and a complete noisy channel characteristic matrix corresponding to the actual channel state matrix obtained using a noise scheduling parameter; the noise scheduling parameter is used to control the noise addition ratio; a correction module, configured to correct the second channel characteristic matrix according to a noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain an updated first channel characteristic matrix; and continue to input the updated first channel characteristic matrix into the denoising network until a preset denoising number of times is reached; A determination module is used to determine a target channel state matrix corresponding to the pilot information based on all the first channel characteristic matrices finally obtained.

[0014] In one possible implementation, the correction module is specifically used to determine the noise error adjustment strength based on the numerical relationship between the noise error value and the noise error threshold; and to correct the second channel characteristic matrix based on the noise error adjustment strength and the noise error value to obtain the latest first channel characteristic matrix.

[0015] In a possible implementation manner, the correction module is further configured to: Determining whether the noise error value is greater than the noise error threshold; If the noise error value is greater than the noise error threshold, calculating a noise error adjustment strength according to the noise error value; If the noise error value is less than or equal to the noise error threshold, the noise error adjustment strength is set to 1.

[0016] In a possible implementation manner, the correction module is further configured to: Substituting the noise error value into the following formula, the noise error adjustment strength is obtained; ; in, Adjust the strength for noise error, is the noise error value, is the remaining number of processing times.

[0017] In a possible implementation manner, the correction module is further configured to: Substituting the noise error adjustment strength and the noise error value into the following formula to obtain the latest first channel characteristic matrix; ; in, is the latest first channel feature matrix, is the second channel feature matrix, is the remaining number of processing times, is the denoising network, are the trainable parameters of the denoising network, is the latest first channel feature matrix obtained when the remaining processing times is t+1.

[0018] In one possible embodiment, the device also includes a training module, which is specifically used to extract the actual channel characteristic matrix of the channel state information at each scale and each direction from the actual channel state matrix; iteratively add noise to the actual channel characteristic matrix according to the noise scheduling parameters to obtain a complete noisy channel characteristic matrix; denoise and correct the complete noisy channel characteristic matrix through the denoising network to obtain a restored channel state matrix corresponding to the complete noisy channel characteristic matrix; update the parameters in the denoising network with the minimum difference between the restored channel state matrix and the actual channel state matrix as the optimization goal; and continue to denoise and correct the complete noisy channel characteristic matrix through the denoising network until a preset iteration stop condition is met.

[0019] In a possible implementation, the training module is further configured to: Inputting the noise scheduling parameters and the actual channel characteristic matrix into the following formula to obtain a complete noise channel characteristic matrix; ; ; in, is the complete noise channel characteristic matrix, is the noise scheduling parameter of the i-th iteration, The number of times noise is added for a preset iteration, is the noise scheduling parameter for the j+1th iteration, is the standard complex Gaussian noise value of the jth iteration that follows the standard normal distribution , I is the covariance matrix of the standard normal distribution, is the actual channel characteristic matrix.

[0020] In a third aspect, an embodiment of the present application further provides an electronic device comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the channel estimation method for wireless communication as described in any one of the first aspects.

[0021] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the channel estimation method for wireless communication as described in any one of the first aspects are executed.

[0022] The embodiment of the present application provides a channel estimation method, device, electronic device and storage medium for wireless communication, the method comprising: extracting the first channel characteristic matrix at each scale and in each direction from the initial channel state matrix contained in the pilot information; for each first channel characteristic matrix, inputting the latest first channel characteristic matrix into the denoising network to obtain the second channel characteristic matrix; correcting the second channel characteristic matrix according to the noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain the latest first channel characteristic matrix; and continuing to input the latest first channel characteristic matrix into the denoising network until a preset number of denoising times is reached; based on all the first channel characteristic matrices finally obtained, determining the target channel state matrix corresponding to the pilot information. The present application performs channel estimation by removing noise and correcting iteratively for multiple times, which can ensure channel estimation performance and improve robustness when the noise level is high or the channel conditions vary greatly. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 A flow chart of a wireless communication channel estimation method provided by an embodiment of the present application is shown; Figure 2 A training flow chart of a denoising network provided in an embodiment of the present application is shown; Figure 3 A schematic structural diagram of a channel estimation device for wireless communication provided in an embodiment of the present application is shown; Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of illustration and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowcharts can be implemented out of sequence, and steps without logical context can be reversed or implemented simultaneously. In addition, those skilled in the art, under the guidance of the contents of this application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0026] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0027] To enable those skilled in the art to utilize the present disclosure, the following embodiments are provided in conjunction with the specific application scenario of "wireless communications." Those skilled in the art will appreciate that the general principles defined herein can be applied to other embodiments and application scenarios without departing from the spirit and scope of this disclosure. While this disclosure is primarily described in the context of "wireless communications," it should be understood that this is merely an exemplary embodiment.

[0028] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0029] The following describes in detail a channel estimation method for wireless communication provided in an embodiment of the present application.

[0030] Reference Figure 1 As shown, it is a flow chart of a channel estimation method for wireless communication provided in an embodiment of the present application. The exemplary steps of the embodiment of the present application are described below: S101 : Extracting a first channel characteristic matrix at each scale and in each direction from an initial channel state matrix included in pilot information.

[0031] In the embodiment of the present application, in a MIMO (Multiple Input Multiple Output) system, the initial channel state matrix (CSI, , where C is a complex set with a size of ) describes the initial channel state between each antenna at the transmitter and each antenna at the receiver. The initial channel state matrix It is a complex matrix. In the Cartesian coordinate system, the distance from the Cartesian coordinate corresponding to each complex element in the initial channel state matrix to the origin represents the channel strength, and the angle between the Cartesian coordinate corresponding to each complex element and the origin represents the channel delay. , where the real part matrix ; , is the imaginary unit.

[0032] The embodiment of the present application uses a two-dimensional discrete wavelet transform (2D-DWT) to transform the real part matrix and the real part matrix of the initial channel state matrix into Decompose into sub-bands of different scales and directions, that is, the first channel feature matrices of different scales and directions.

[0033] Here, the first channel feature matrix includes an approximation coefficient matrix, a vertical detail coefficient matrix, a horizontal detail coefficient matrix, and a diagonal detail coefficient matrix. (1) The approximation coefficient matrix is used to represent the low-frequency part of the channel state matrix, that is, the smoothness or overall trend of the channel. They capture the main characteristics of the channel but remove the detail information. (2) The vertical detail coefficient matrix is used to represent the high-frequency part of the channel state matrix in the vertical direction, that is, the rapid changes or edge information of the channel in the vertical direction. They capture the detailed characteristics of the channel in the vertical direction. (3) The horizontal detail coefficient matrix is used to represent the high-frequency part of the channel state matrix in the horizontal direction, that is, the rapid changes or edge information of the channel in the horizontal direction. They capture the detailed characteristics of the channel in the horizontal direction. (4) The diagonal detail coefficient matrix is used to represent the high-frequency part of the channel state matrix in the diagonal direction, that is, the rapid changes or edge information of the channel in the diagonal direction. They capture the detailed characteristics of the channel in the diagonal direction.

[0034] Specifically, the following steps are performed to transform the real part matrix or Decompose into the first channel feature matrix of different scales and directions: First, for the real matrix or Each line Perform row filtering to obtain the intermediate approximate coefficients and intermediate detail coefficients :

[0035] in, is the number of rows of the real or imaginary matrix, is the real matrix or The number of columns, is the intermediate approximate coefficient The value of the mth row and kth column in is the real matrix or , is the intermediate detail coefficient The value of the mth row and kth column in are the low-pass filter coefficients associated with the selected wavelet, High-pass filter coefficients associated with the selected wavelet.

[0036] Then, for the intermediate approximation coefficient and intermediate detail coefficients Each column Applying one-dimensional DWT, we get the first channel feature matrix of different scales and directions (its size is ):

[0037] in, is the approximate coefficient matrix, is the vertical detail coefficient matrix, is the horizontal detail coefficient matrix, is the diagonal detail coefficient matrix.

[0038] S102: For each first channel feature matrix, input the latest first channel feature matrix into a denoising network to obtain a second channel feature matrix.

[0039] Among them, the denoising network is trained based on the actual channel state matrix and the complete noise channel feature matrix corresponding to the actual channel state matrix obtained using the noise scheduling parameters; the noise scheduling parameters are used to control the noise addition ratio.

[0040] Reference Figure 2 As shown, it is a training flow chart of the denoising network provided in an embodiment of the present application. The exemplary steps of the embodiment of the present application are described below: S201 : Extracting an actual channel characteristic matrix of the channel state information at each scale and in each direction from the actual channel state matrix.

[0041] In the embodiment of the present application, the method of extracting the actual channel characteristic matrix from the actual channel state matrix is the same as the process of extracting the first channel characteristic matrix from the initial channel state matrix, which will not be repeated here.

[0042] The actual channel state matrix includes the actual state of the channel.

[0043] S202: Iteratively add noise to the actual channel characteristic matrix according to the noise scheduling parameters to obtain a complete noise channel characteristic matrix.

[0044] In the embodiment of the present application, during the forward process (training phase) of the diffusion model, Gaussian noise is gradually added to the inter-channel characteristic matrix until it becomes completely random noise, that is, a completely noisy channel characteristic matrix is obtained.

[0045] Here, in this process, the diffusion model simulates the degradation process of the data by gradually adding noise until the data becomes completely noisy. This process can be regarded as building a Markov chain in which each state depends only on the previous state. Since the true distribution is unknown, we use the state probability predicted by the diffusion model To approximate the true state profile .

[0046] ; ; in, is the total number of diffusion steps (typical value is 200-1000 steps, which controls the rhythm of noise addition and can also be called the preset number of iterative noise additions), For probability.

[0047] In order to To obtain a Gaussian white noise distribution, we use two specific scalars (hyperparameters of the diffusion model) and , are the trainable parameters of the denoising network in the diffusion model.

[0048] The calculation formula is: ; The following formula can be obtained by cumulatively superimposing the above calculation formulas: ; ; in, is the complete noise channel characteristic matrix, is the noise scheduling parameter of the i-th iteration, The number of times noise is added for a preset iteration, is the noise scheduling parameter for the j+1th iteration, is the standard complex Gaussian noise value of the jth iteration that follows the standard normal distribution , I is the covariance matrix of the standard normal distribution, is the actual channel characteristic matrix.

[0049] S203 , denoising and correcting the completely noisy channel characteristic matrix through a denoising network to obtain a restored channel state matrix corresponding to the completely noisy channel characteristic matrix.

[0050] In an embodiment of the present application, for each completely noisy channel characteristic matrix, the latest completely noisy channel characteristic matrix is input into the denoising network to obtain a third channel characteristic matrix; based on the noise error value between the latest completely noisy channel characteristic matrix and the third channel characteristic matrix, the third channel characteristic matrix is corrected to obtain the latest completely noisy channel characteristic matrix; and the latest completely noisy channel characteristic matrix continues to be input into the denoising network until a preset number of denoising cycles is reached, at which point the latest completely noisy channel characteristic matrix is determined as the restored channel state matrix.

[0051] S204, with the minimum difference between the restored channel state matrix and the actual channel state matrix as the optimization goal, update the parameters in the denoising network; and continue to denoise and correct the complete noisy channel feature matrix through the denoising network until the preset iteration stop condition is met.

[0052] In the embodiment of the present application, in the forward process of the diffusion model, the denoising network is used to learn how to predict noise at any time and optimize the relevant trainable parameters of the diffusion model (including the parameters in the denoising network).

[0053] Here, existing solutions generally use end-to-end neural networks. This technical solution only uses an end-to-end denoising network to fit the noise (S201 to S204) and uses a diffusion model framework to solve the channel estimation problem, so that the entire channel estimation method is robust and can adapt to changing environmental information.

[0054] In addition, in the forward process, CNN can be used to fit noise, such as a fully connected network, U-net, and the same effect can be achieved by increasing or decreasing the number of network layers.

[0055] Other generative networks such as GAN and VAE can also achieve similar effects.

[0056] S103. According to the noise error value between the first channel characteristic matrix and the second channel characteristic matrix, the second channel characteristic matrix is corrected to obtain the latest first channel characteristic matrix; and the latest first channel characteristic matrix is continuously input into the denoising network until a preset denoising number is reached.

[0057] In an embodiment of the present application, the noise error adjustment strength is determined based on the numerical relationship between the noise error value and the noise error threshold; the second channel characteristic matrix is corrected based on the noise error adjustment strength and the noise error value to obtain the latest first channel characteristic matrix.

[0058] Specifically, determining the noise error adjustment strength according to the numerical relationship between the noise error value and the noise error threshold includes: Step 1: Determine whether the noise error value is greater than the noise error threshold.

[0059] In the embodiment of the present application, the noise error threshold is used to distinguish whether the noise in the latest first channel characteristic matrix is in a high noise area or a low noise area.

[0060] Step 2: If the noise error value is greater than the noise error threshold, the noise error adjustment strength is calculated according to the noise error value.

[0061] In the embodiment of the present application, if the noise error value is greater than the noise error threshold, that is, the noise in the latest first channel characteristic matrix is in a high noise area, in order to ensure the denoising effect, the correction amplitude is reduced, that is, the noise error adjustment strength is calculated based on the noise error value. The noise error adjustment strength is less than 1.

[0062] Here, the noise error adjustment strength is calculated according to the noise error value, including: Substitute the noise error value into the following formula to obtain the noise error adjustment strength; ; in, Adjust the strength for noise error, is the noise error value, is the remaining number of processing times.

[0063] Step 3: If the noise error value is less than or equal to the noise error threshold, the noise error adjustment strength is set to 1.

[0064] In an embodiment of the present application, if the noise error value is less than or equal to the noise error threshold, that is, the noise in the latest first channel feature matrix is in a low noise area, in order to ensure the denoising efficiency, the noise error adjustment intensity is set to 1 to ensure complete noise transmission.

[0065] Specifically, the second channel characteristic matrix is corrected according to the noise error adjustment strength and the noise error value to obtain the latest first channel characteristic matrix, including: Substitute the noise error adjustment strength and the noise error value into the following formula to obtain the latest first channel feature matrix; ; in, is the latest first channel feature matrix, is the second channel feature matrix, is the remaining number of processing times, is the denoising network, are the trainable parameters of the denoising network, is the latest first channel feature matrix obtained when the remaining processing times is t+1.

[0066] Here, the remaining processing times are used to select the appropriate diffusion model time step using the signal-to-noise ratio (SNR) information to optimize the channel estimation efficiency.

[0067] S104: Determine a target channel state matrix corresponding to the pilot information based on all the first channel characteristic matrices finally obtained.

[0068] In the embodiment of the present application, all the first channel feature matrices finally obtained are subjected to 2D-DWT inverse transformation to convert the estimation results from the angular domain back to the spatial domain to obtain the target channel state matrix.

[0069] Compared to Gaussian random sampling in traditional diffusion models, this application uses a dynamic cold diffusion method for sampling (S102 to S104), gradually applying the learned noise prediction model for denoising. This sampling strategy considers prediction error information when processing noise, is more tolerant to ambient noise, and avoids the traditional random resampling process. Each state is adjusted until a new, realistic channel state is reconstructed. This process uses a deterministic inverse process until the desired signal-to-noise ratio (SNR) level is reached.

[0070] In this embodiment, the unique inverse process (S102 to S104) and forward process (S201 to S204) are primarily used to accurately estimate channel state information in the presence of Gaussian noise. Compared to traditional channel estimation methods, the diffusion model (DM) offers the advantage of maintaining high estimation accuracy in noisy environments. By reducing the number of steps in the inverse process, computational latency is effectively reduced, significantly improving the timeliness of channel estimation, which is particularly critical in rapidly changing wireless environments.

[0071] Specifically, the diffusion model, through its Markov chain model, treats channel estimation as a generative task of latent variables, approximating it at each step via a parameterized Gaussian transfer function. Compared to other generative models, the diffusion model (DM) exhibits better asymptotic and robust properties in gradually approximating the true channel model estimate. This approximation capability enables DM to provide consistent and efficient channel estimation under varying signal-to-noise ratio conditions, particularly in massive MIMO systems with limited resource and computational requirements. Traditional machine learning channel estimation models, on the other hand, perform poorly in scenarios with signal-to-noise ratios outside the training set.

[0072] The design of the diffusion model not only considers signal reconstruction accuracy but also incorporates a low-latency computational strategy, enabling it to outperform traditional methods in large-scale MIMO systems. By combining subband learning with channel distribution in the sparse angular domain, DM achieves an effective balance between resource consumption and computational efficiency, providing a practical and highly innovative channel estimation solution for large-scale MIMO systems.

[0073] In summary, the application of DM in MIMO channel estimation not only improves the CSI recovery quality, but also optimizes the system calculation and resource efficiency. Its innovation and technical effects have far-reaching impact and application potential in the context of current B5G communication technology.

[0074] Based on the same inventive concept, an embodiment of the present application also provides a channel estimation device for wireless communication corresponding to the channel estimation method for wireless communication. Since the principle of solving the problem by the device in the embodiment of the present application is similar to the channel estimation method for wireless communication in the embodiment of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0075] Reference Figure 3 FIG. 1 is a schematic diagram of a channel estimation device for wireless communication provided in an embodiment of the present application, wherein the channel estimation device for wireless communication includes: An extraction module 301 is configured to extract a first channel characteristic matrix at each scale and in each direction from an initial channel state matrix included in the pilot information; An input module 302 is configured to input the latest first channel characteristic matrix into a denoising network for each first channel characteristic matrix to obtain a second channel characteristic matrix; the denoising network is trained based on an actual channel state matrix and a complete noisy channel characteristic matrix corresponding to the actual channel state matrix obtained using a noise scheduling parameter; the noise scheduling parameter is used to control the noise addition ratio; a correction module 303 configured to correct the second channel characteristic matrix according to a noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain an updated first channel characteristic matrix; and continuously input the updated first channel characteristic matrix into the denoising network until a preset number of denoising cycles is reached; The determination module 304 is configured to determine a target channel state matrix corresponding to the pilot information based on all the first channel characteristic matrices finally obtained.

[0076] like Figure 4As shown, an electronic device 400 provided in an embodiment of the present application includes: a processor 401, a memory 402 and a bus, wherein the memory 402 stores machine-readable instructions executable by the processor 401. When the electronic device is running, the processor 401 communicates with the memory 402 through the bus, and the processor 401 executes the machine-readable instructions to perform the steps of the channel estimation method for wireless communication as described above.

[0077] Specifically, the memory 402 and the processor 401 can be general-purpose memories and processors, which are not specifically limited here. When the processor 401 runs the computer program stored in the memory 402, the channel estimation method for wireless communication can be executed.

[0078] Corresponding to the above-mentioned channel estimation method for wireless communication, an embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned channel estimation method for wireless communication are executed.

[0079] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0080] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0081] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0082] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the information processing method described in each embodiment of this application. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0083] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A channel estimation method for wireless communication, characterized in that: The method comprises: Extracting a first channel characteristic matrix at each scale and in each direction from an initial channel state matrix included in the pilot information; For each first channel characteristic matrix, inputting the latest first channel characteristic matrix into a denoising network to obtain a second channel characteristic matrix; the denoising network is trained based on an actual channel state matrix and a complete noisy channel characteristic matrix corresponding to the actual channel state matrix obtained using a noise scheduling parameter; the noise scheduling parameter is used to control the noise addition ratio; Correcting the second channel characteristic matrix according to a noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain an updated first channel characteristic matrix; and continuously inputting the updated first channel characteristic matrix into the denoising network until a preset number of denoising cycles is reached; Based on all the first channel characteristic matrices finally obtained, a target channel state matrix corresponding to the pilot information is determined.

2. The wireless communication channel estimation method according to claim 1, wherein: The correcting the second channel characteristic matrix according to the noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain the latest first channel characteristic matrix includes: determining a noise error adjustment strength according to a numerical relationship between the noise error value and a noise error threshold; The second channel characteristic matrix is corrected according to the noise error adjustment strength and the noise error value to obtain the latest first channel characteristic matrix.

3. The wireless communication channel estimation method according to claim 2, wherein: The determining the noise error adjustment strength according to the numerical relationship between the noise error value and the noise error threshold includes: Determining whether the noise error value is greater than the noise error threshold; If the noise error value is greater than the noise error threshold, calculating a noise error adjustment strength according to the noise error value; If the noise error value is less than or equal to the noise error threshold, the noise error adjustment strength is set to 1.

4. The wireless communication channel estimation method according to claim 3, wherein: Calculating the noise error adjustment strength according to the noise error value includes: Substituting the noise error value into the following formula, the noise error adjustment strength is obtained; ； in, Adjust the strength for noise error, is the noise error value, is the remaining number of processing times.

5. The wireless communication channel estimation method according to claim 2, wherein: The step of correcting the second channel characteristic matrix according to the noise error adjustment strength and the noise error value to obtain the latest first channel characteristic matrix includes: Substituting the noise error adjustment strength and the noise error value into the following formula to obtain the latest first channel characteristic matrix; ； in, is the latest first channel feature matrix, is the second channel characteristic matrix, is the remaining number of processing times, is the denoising network, are the trainable parameters of the denoising network, is the latest first channel feature matrix obtained when the remaining processing times is t+1.

6. The wireless communication channel estimation method according to any one of claims 1 to 5, characterized in that: The denoising network is trained by the following steps: Extracting the actual channel feature matrix of the channel state information at each scale and in each direction from the actual channel state matrix; Iteratively adding noise to the actual channel characteristic matrix according to the noise scheduling parameters to obtain a complete noise channel characteristic matrix; After denoising and correcting the complete noisy channel characteristic matrix through the denoising network, a restored channel state matrix corresponding to the complete noisy channel characteristic matrix is obtained; The parameters in the denoising network are updated with the minimum difference between the restored channel state matrix and the actual channel state matrix as the optimization goal; and the completely noisy channel characteristic matrix is continuously denoised and corrected through the denoising network until a preset iteration stop condition is met.

7. The wireless communication channel estimation method according to claim 6, wherein: The iteratively adding noise to the actual channel characteristic matrix according to the noise scheduling parameter to obtain a complete noise channel characteristic matrix includes: Inputting the noise scheduling parameters and the actual channel characteristic matrix into the following formula to obtain a complete noise channel characteristic matrix; ； ； in, is the complete noise channel characteristic matrix, is the noise scheduling parameter of the i-th iteration, The number of times noise is added for a preset iteration, is the noise scheduling parameter for the j+1th iteration, is the standard complex Gaussian noise value of the jth iteration that follows the standard normal distribution , I is the covariance matrix of the standard normal distribution, is the actual channel characteristic matrix.

8. A channel estimation device for wireless communication, characterized in that: The device comprises: An extraction module, configured to extract a first channel characteristic matrix at each scale and in each direction from an initial channel state matrix included in the pilot information; An input module is configured to input the latest first channel characteristic matrix into a denoising network for each first channel characteristic matrix to obtain a second channel characteristic matrix; the denoising network is trained based on an actual channel state matrix and a complete noisy channel characteristic matrix corresponding to the actual channel state matrix obtained using a noise scheduling parameter; the noise scheduling parameter is used to control the noise addition ratio; a correction module, configured to correct the second channel characteristic matrix according to a noise error value between the first channel characteristic matrix and the second channel characteristic matrix to obtain an updated first channel characteristic matrix; and continue to input the updated first channel characteristic matrix into the denoising network until a preset denoising number of times is reached; A determination module is used to determine a target channel state matrix corresponding to the pilot information based on all the first channel characteristic matrices finally obtained.

9. An electronic device, characterized in that: include: A processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of the channel estimation method for wireless communication according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the channel estimation method for wireless communication according to any one of claims 1 to 7 are executed.

Images