Self-adaptive underwater acoustic channel estimation method based on discrete Hartley transform

Abstract

The invention provides a self-adaptive underwater acoustic channel estimation method based on discrete Hartley transform. The method is used for improving the convergence speed, the calculation efficiency and the robustness of underwater acoustic channel estimation. The method has the advantages that the discrete Hartley transform (DHT) mechanism is combined with the self-adaptive normalized step length LMS method, real-time adjustment of the self-adaptive step length is achieved by conducting Hartley transform and dynamic power estimation updating on input signals, and the convergence speed and stability of the algorithm in the non-stationary underwater acoustic channel environment are effectively improved. According to the method, the low calculation complexity can be kept under the condition that the power fluctuation of the input signal is large, and the self-adaptability and the anti-interference capability of the filter are remarkably improved. Through real-time adaptive adjustment under different underwater acoustic channel conditions, efficient tracking and accurate estimation of the underwater acoustic channel are realized, the real-time performance and accuracy of an underwater acoustic communication system are further enhanced, and reliable technical support is provided for signal processing in a complex underwater environment.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater acoustic communication, and in particular to an adaptive underwater acoustic channel estimation method based on discrete Hartley transform. Background Art

[0002] In the field of underwater acoustic communication, channel estimation and noise suppression are key issues to improve communication reliability and accuracy. Based on the optimization model of sparse channel response, existing underwater acoustic channel estimation methods mainly rely on traditional adaptive filter design, such as Normalized Least Mean Squares (NLMS) and Data-Reuse Recursive Least-Squares (DRRLS). These algorithms take advantage of signal sparsity and optimize the filter coefficients iteratively to enhance the adaptability to channel characteristics. For example, the sparse estimation algorithm based on normalized subband adaptive filters (see "Sparsity-aware normalized subband adaptive filters with jointly optimized parameters") has achieved remarkable results in sparse signal processing. By jointly optimizing the subband coefficients, it realizes the effective estimation of sparse channels. Similarly, the data reuse recursive least squares algorithm (see "Data-Reuse Recursive Least-Squares Algorithms") significantly improves the convergence speed on the basis of signal reuse and is suitable for scenarios with small channel changes. However, these algorithms based on fixed step size and conventional filter structure have limited performance under complex and changeable underwater acoustic channel conditions and are difficult to cope with the complex interference of underwater noise and non-stationary signals.

[0003] The complexity of the underwater acoustic channel is mainly reflected in its strong non-stationary characteristics and significant multipath effects, which greatly increases the difficulty of accurately modeling and estimating the underwater acoustic channel. The traditional adaptive filter based on fixed step size is difficult to cope with the dynamic characteristics of the underwater acoustic channel, resulting in insufficient estimation of the channel response. For example, in a complex and dynamically changing underwater environment, the step size of the traditional method is difficult to keep up with the rapid changes in the time-varying characteristics of the channel, which limits its estimation accuracy. At the same time, the noise and multipath effects in the underwater acoustic channel also make the estimation of the channel response face great challenges. Under the influence of the multipath effect, the channel response may present sparse or highly sparse structural features. Therefore, adaptive adjustment of the channel coefficients and power estimation are of great significance in improving the accuracy and robustness of the channel estimation. To this end, the estimation method of the underwater acoustic channel needs to be carefully optimized in terms of adaptive step size and channel power estimation to cope with the complexity of the underwater acoustic environment.

[0004] Although traditional adaptive filtering algorithms have been used in underwater acoustic channel estimation, there is an obvious trade-off between convergence accuracy and convergence speed. In order to cope with the complexity of underwater acoustic channels, fast convergence and accurate estimation are particularly important in dynamic underwater environments, and traditional adaptive methods are difficult to meet both requirements at the same time. To this end, an underwater acoustic channel estimation method based on sparse domain inverse power normalization is proposed. The channel response is mapped to the sparse domain through discrete Hartley transform (DHT), and combined with an adaptive step size adjustment mechanism, efficient estimation of the channel response is achieved. At the same time, an inverse power normalization strategy is introduced to finely adjust the filter gain through power estimation to stabilize the estimation effect of the channel response, thereby significantly improving the channel estimation accuracy and robustness in complex underwater acoustic environments. Summary of the invention

[0005] In order to effectively solve the problem of complex and changeable underwater acoustic channels, the present invention proposes an adaptive underwater acoustic channel estimation method based on discrete Hartley transform (DHT). The invention utilizes the discrete Hartley transform framework, combined with an adaptive inverse power normalization mechanism, to achieve high-precision channel estimation in complex underwater environments by dynamically adjusting the step size and real-time power estimation. By introducing a sparse domain channel representation strategy and adopting an inverse power normalization method, it is possible to effectively balance the convergence speed and stability of channel estimation, and ultimately accurately reconstruct the underwater acoustic channel response, significantly improving the robustness and performance of the overall communication system;

[0006] In order to solve the problem that the classical underwater acoustic channel estimation algorithm is difficult to balance the amount of calculation, convergence speed and accuracy, the specific implementation steps of the present invention are as follows:

[0007] (1) Set the parameter μ0 as the initial step size, set λ as the smoothing coefficient, and H as the discrete Hartley transform matrix. Initialize the weight vector w = 0, the instantaneous power vector p = 0, and the mean square error vector MSD = 0;

[0008] (2) Given the input signal x, the expected signal d and the initial weight h, repeat the following i-th iteration (i = 1, 2, ..., N):

[0009] Calculate the current input vector x v , where xv = [x i ,x i-1 ,...,x i-L+1 ];

[0010] Computes the discrete Hartley transform of the input vector

[0011] Calculate the smoothed power p, where p = λp + (1-λ) |u| 2 ,|u| 2 represents the square modulus of each element in u;

[0012] Compute the inverse diagonal matrix H inv =diag(1 / p);

[0013] Calculate the normalization factor f n , where f n =u T H inv u+10 -6 ;

[0014] Calculate the output y i =u T w and calculate the error e = d i -y i ;

[0015] Calculate the normalized step size factor μ = μ0 / f n ;

[0016] Update weight vector w = w + μH inv ue * ;

[0017] Calculate and record the mean square error MSD i =|wh| 2 / |h| 2 .

[0018] The present invention provides an adaptive underwater acoustic channel estimation method based on discrete Hartley transform. It has the following beneficial effects:

[0019] The present invention proposes an adaptive underwater acoustic channel estimation method based on discrete Hartley transform, which is used to improve the convergence speed, computational efficiency and robustness of underwater acoustic channel estimation. The beneficial effects are embodied in: the present invention utilizes the discrete Hartley transform (DHT) mechanism combined with the adaptive normalized step size LMS method, and realizes the real-time adjustment of the adaptive step size by performing Hartley transform and dynamic power estimation update on the input signal, which effectively improves the convergence speed and stability of the algorithm in a non-stationary underwater acoustic channel environment. The method can maintain a low computational complexity when the input signal power fluctuates greatly, and significantly improves the adaptability and anti-interference ability of the filter. Through real-time adaptive adjustment under different underwater acoustic channel conditions, the present invention realizes efficient tracking and accurate estimation of the underwater acoustic channel, further enhances the real-time and accuracy of the underwater acoustic communication system, and provides reliable technical support for signal processing in complex underwater environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1The learning curve comparison results of the method of the present invention (new Discrete Hartley Transform Least Mean Squares, nDHTLMS) and the traditional minimum mean square error method, the normalized least mean square error algorithm (Normalized Least-Mean-Squares, NLMS), and the data reuse recursive least squares algorithm (Data-Reuse Recursive Least-Squares, DR-RLS) are shown.

[0021] Figure 2 This is a comparison chart of the calculation time of the traditional methods LMS, NLMS, DR-RLS and the method of the present invention (nDHTLMS). DETAILED DESCRIPTION

[0022] From the experimental results, the proposed adaptive filtering algorithm based on discrete Hartley transform (nDHTLMS) shows significant advantages in terms of convergence speed, robustness and computational efficiency. Figure 1 From the learning curve, we can see that nDHTLMS is compared with traditional LMS, NLMS and DR-RLS algorithms under the conditions of signal-to-noise ratio (SNR) of 10dB and 20dB. Whether in low SNR environment (10dB) or high SNR environment (20dB), the mean square error (MSD) convergence speed of nDHTLMS algorithm is significantly faster than that of traditional algorithms, which indicates that it has superior real-time tracking capability. This shows that the algorithm can quickly converge to a stable state and is suitable for dynamically changing underwater acoustic channel environments.

[0023] Combination Figure 1 and Figure 2 It can be observed that although the DR-RLS algorithm has a relatively fast convergence speed, its computational complexity is high and its running time is significantly higher than other algorithms, making it unsuitable for real-time applications. Therefore, the nDHTLMS algorithm achieves the goals of fast convergence and low steady-state error while maintaining low computational complexity, showing significant advantages in underwater acoustic channel estimation. Although the LMS and NLMS algorithms are computationally simple, their convergence speed and steady-state error are not as good as nDHTLMS. The running time of the nDHTLMS algorithm is close to that of LMS and NLMS, indicating that its computational complexity is low and it is suitable for real-time application scenarios, especially in resource-constrained underwater acoustic communication environments. This result shows that the nDHTLMS algorithm achieves a high convergence speed and estimation accuracy while ensuring computational efficiency, and has broad application potential.

[0024] The adaptive step size adjustment mechanism of the nDHTLMS algorithm during the iteration process plays a key role in improving its performance. By introducing discrete Hartley transform and dynamic power estimation, the algorithm can adjust the step size in each iteration to adapt it to the changing characteristics of the channel, thereby avoiding the divergence problem caused by too large a step size and reducing the slow convergence caused by too small a step size. This feature enables the nDHTLMS algorithm to achieve fast response in dynamic channels, thereby performing well in underwater acoustic channel estimation tasks, and is particularly suitable for scenarios with high real-time requirements.

[0025] In summary, the nDHTLMS algorithm has obvious advantages in channel estimation performance. It can be seen from the experimental data that the algorithm not only has faster convergence speed and lower steady-state error in low signal-to-noise ratio environments, but also maintains low computational complexity and running time. Compared with the traditional LMS and NLMS algorithms, nDHTLMS fully utilizes the advantages of discrete Hartley transform and achieves a better balance: it not only improves the convergence speed of the adaptive filtering algorithm, but also maintains computational efficiency, providing a reliable technical solution for real-time estimation of underwater acoustic channels.

[0026] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0027] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. An adaptive underwater acoustic channel estimation method based on discrete Hartley transform, characterized in that: The method includes the following: S1: Set the parameter μ0 as the initial step size, set λ as the smoothing coefficient, H as the discrete Hartley transform matrix, initialize the weight vector w=0, the instantaneous power vector p=0 and the mean square error vector MSD=0; S2: Given the input signal x, the desired signal d and the initial weight h, repeat the following i-th iteration (i=1,2,...,N): S3: Calculate the current input vector x v , where xv = [x i ,x i-1 ,...,x i-L+1 ]; S4: Calculate the input vector after discrete Hartley transform S5: Calculate the smoothed power p, where p = λp + (1-λ) |u| 2 ,|u| 2 represents the square modulus of each element in u; S6: Calculate the inverse diagonal matrix H inv =diag(1 / p); S7: Calculate the normalization factor f n , where f n =u T H inv u+10 -6 ; S8: Calculate output y i =u T w and calculate the error e = d i -y i ; S9: Calculate the normalized step size factor μ = μ0 / f n ; S10: Update weight vector w=w+μH inv ue * ; S11: Calculate and record the mean square error MSD i =|wh| 2 / |h| 2 .

2. The method for adaptive underwater acoustic channel estimation based on discrete Hartley transform according to claim 1, characterized in that: High-precision channel estimation is achieved in a time-varying underwater acoustic channel environment. This method introduces an inverse power normalization strategy in the sparse domain, adaptively adjusts the step size to adapt to the dynamic changes of the channel, and combines the discrete Hartley transform (DHT) matrix to enhance the ability to capture sparse features, thereby improving the convergence speed and estimation accuracy of the algorithm in a complex underwater acoustic environment.

Images