A robust estimation method and device for underwater acoustic channel resistant to ice sheet noise
By using the algorithm of Lawson norm and RLS error cost function in polar ice sub-acoustic water acoustic channel, the problem of inaccurate estimation of small energy multipath amplitude in the traditional method in the impulse noise environment is solved, and the robustness and estimation accuracy are improved, and the accuracy of estimation is achieved. Stable estimates of subic channels.
Patent Information
- Application Number
- CN202311043072.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-18
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2043-08-18
AI Technical Summary
When facing impulse noise in a polar ice submersible water acoustic environment, traditional channel estimation algorithm cannot achieve accurate estimation of small energy multipath amplitude, and is poorly robust.
Using an algorithm based on Lawson norm and RLS error cost function, the robustness of impulse noise is improved and the estimation accuracy of amplitude and delay is improved by constructing feature matrix and iterative formula.
The small amplitude multipath robust estimation of subic ice channels is realized, which improves the robustness to impulse noise, and has good steady-state error and convergence rate. The experimental data verifies the effectiveness of the method.
Smart Images

Figure CN117544453B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of polar acoustics and underwater acoustic signal processing, and relates to a method and device for robustly estimating an underwater acoustic channel under ice and resistant to ice field noise. Background Art
[0002] The climate in the Arctic is extremely cold and has a very complex underwater acoustic environment. Due to friction and fractures between ice layers, pulse noise of abnormal amplitude is generated. The generation of pulse noise is random and different from Gaussian noise in open waters. Its second-order moment and higher-order moment are infinite.
[0003] Most traditional channel estimation algorithms are aimed at the background interference of Gaussian noise in open waters. For Gaussian noise, the two-norm has the optimal solution to its constraints. Therefore, conventional channel estimation algorithms have good suppression capabilities for Gaussian noise. However, due to the statistical characteristics of impulse noise, the performance of traditional two-norm-based channel estimation algorithms is greatly reduced. To address this problem, many channel estimation algorithms based on norm penalty have been proposed. This algorithm expands the two-norm space to the one-norm and zero-norm space, thereby suppressing impulse noise. Although the channel estimation algorithm based on norm penalty improves the robustness to impulse noise, it cannot accurately estimate the amplitude of small energy multipath in underwater acoustic systems. Summary of the invention
[0004] The purpose of the present invention is to propose a robust estimation method and device for underwater acoustic channels that are resistant to ice field noise, as the existing underwater acoustic channel estimation methods are unable to achieve accurate estimation of the small energy multipath amplitude in the underwater acoustic system due to poor impulse noise suppression effect. The method and device are based on the Lawson norm and the RLS error cost function for estimation, which can effectively improve the robustness of polar underwater channels to impulse noise and improve the accuracy of amplitude and delay.
[0005] In order to achieve the above-mentioned object, the present invention adopts the following technical scheme. According to the first aspect of the present invention, the present invention claims protection for a robust estimation method of underwater acoustic channel under ice that is resistant to ice sheet noise, comprising:
[0006] Step 1: construct a cost function of the Lawson-RLS algorithm, obtain an estimated weight vector and an estimated error according to the cost function, construct a feature matrix based on the Lawson norm, and preset parameters of the feature matrix through the transmitted signal and the received signal;
[0007] Step 2: Set the initial value of the characteristic matrix, then solve the gain matrix to obtain the gain vector;
[0008] Step 3: Obtaining the iterative formula of the Lawson-RLS algorithm according to the gain vector and the characteristic matrix;
[0009] Step 4: Update the characteristic matrix, and continue to repeat the above steps 1-3 until step 3 meets the iteration number condition, and obtain the estimated values of the channel delay and amplitude.
[0010] Furthermore, the step 1 specifically includes:
[0011] Using the weight vector of multi-dimensional unknown hydroacoustic sparse system;
[0012] Based on time The input signal and ambient noise are obtained at The output of the sparse hydroacoustic system at the moment;
[0013] The cost function is constructed using the Lawson norm of the tap coefficient vector and the RLS function of the error;
[0014] According to the cost function, The estimated weight vector at time The estimated error of the moment;
[0015] Based on the Lawson norm, we get Output values of different multipath channels at different times;
[0016] The individual items of the characteristic matrix are operated according to the gradient descent theory to obtain a transition matrix.
[0017] Furthermore, the step 2 specifically includes:
[0018] Setting the initial value of the characteristic matrix and then solving the gain matrix;
[0019] The initial value of the characteristic matrix is set to 0, and the solution of the tap coefficients of the adaptive filter is obtained;
[0020] Define the weighted complex autocorrelation matrix and solve the gain vector according to the matrix inversion lemma.
[0021] Furthermore, the step 3 specifically includes:
[0022] According to the gain vector, characteristic matrix, The estimated error at the moment and the transition matrix are used to obtain the iterative formula of the Lawson-RLS algorithm.
[0023] Furthermore, the step 4 specifically includes:
[0024] The iteration number condition is set to the length of the received signal, using the prior information of the transmitted signal and the received signal.
[0025] According to a second aspect of the present invention, the present invention claims protection for a robust estimation device for underwater acoustic channels resistant to ice field noise, comprising:
[0026] a memory for storing non-transitory computer-readable instructions; and
[0027] A processor is used to run the computer-readable instructions so that the processor implements the robust estimation method of an underwater acoustic channel resistant to ice sheet noise when executing the instructions.
[0028] The present invention seeks to protect a method and device for robust estimation of underwater acoustic channels resistant to ice sheet noise, including: 1. Constructing a cost function of the Lawson-RLS algorithm, obtaining an estimated weight vector and an estimated error based on the cost function, constructing a characteristic matrix based on the Lawson norm, and presetting parameters of the characteristic matrix through transmitting and receiving signals; 2. Setting the initial value of the characteristic matrix, solving the gain matrix, and obtaining a gain vector; 3. Obtaining an iterative formula for the estimated weight vector based on the gain vector and the characteristic matrix; updating the characteristic matrix, and continuing to repeat 1 to 3 above until the number of iterations is satisfied, to obtain the estimated values of the channel delay and amplitude. It realizes the robust estimation of small-amplitude multipath in the underwater channel, improves the robustness against pulse noise, and the algorithm has good steady-state error and convergence rate. The experimental data verifies the effectiveness of the method.
[0029] Beneficial effects:
[0030] The present invention proposes a robust estimation method and device for underwater acoustic channel resistant to ice sheet noise, which has the following beneficial effects compared with the existing underwater acoustic channel estimation method and device:
[0031] 1. The channel estimation method and device can be used for amplitude and delay estimation of polar subglacial channels;
[0032] 2. The channel estimation method and device achieve robust estimation of small amplitude multipath in the under-ice channel;
[0033] 3. The method and device improve robustness against impulse noise;
[0034] 4. The method and device have good steady-state error and convergence rate, and the experimental data verify the effectiveness of the method. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 This is a working flow chart of a robust estimation method for an underwater acoustic channel resistant to ice sheet noise according to the present invention;
[0036] Figure 2 The present invention provides a robust estimation method for underwater acoustic channels under ice and is resistant to ice noise. The channel estimation is based on the Lawson norm and the RLS error function. The traditional RLS and -Comparison of MSD curves of RLS;
[0037] Figure 3 It is an MSD curve diagram of the channel estimation based on the Lawson norm and RLS error function of the robust estimation method of underwater acoustic channel resistant to ice field noise of the present invention and the comparison with the existing other norm-based channel estimation;
[0038] Figure 4 A schematic diagram of polar ice cracking noise collected during an Arctic scientific expedition involved in the specific implementation of a robust estimation method for an underwater acoustic channel resistant to ice field noise according to the present invention;
[0039] Figure 5 It is a schematic diagram of the channel estimation results based on the Lawson norm and RLS error function after adding polar ice crack noise when implementing a robust estimation method for an underwater acoustic channel resistant to ice field noise according to the present invention. DETAILED DESCRIPTION
[0040] The specific implementation of a method and device for robust estimation of an underwater acoustic channel resistant to ice field noise according to the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0041] Example 1
[0042] According to the first embodiment of the present invention, referring to the attached Figure 1 The present invention claims a robust estimation method of an underwater acoustic channel resistant to ice sheet noise, comprising:
[0043] Step 1: Construct the cost function of the Lawson-RLS algorithm, obtain the estimated weight vector and estimated error based on the cost function, and construct the feature matrix based on the Lawson norm , through the transmission signal and the reception signal, the characteristic matrix Preset parameters;
[0044] Step 2: Set the feature matrix The initial value is , then the gain matrix Solve and get the gain vector ;
[0045] Step 3: Based on the gain vector and the feature matrix Obtaining an iterative formula for the Lawson-RLS algorithm;
[0046] Step 4: Update the feature matrix , continue to repeat the above steps 1-3 until step 3 meets the iteration number condition, and obtain the estimated values of channel delay and amplitude.
[0047] Furthermore, the step 1 specifically includes:
[0048] use The weight vector of the unknown hydroacoustic sparse system of dimension;
[0049] Based on time The input signal x(l) and the ambient noise n(l) at the time of , combined with the weight vector of the unknown underwater acoustic sparse system, are obtained The output d(l) of the sparse hydroacoustic system at time t;
[0050] The Lawson norm and estimation error of the tap coefficient vector The RLS function constructs the cost function;
[0051] Among them, the estimated error By The output d(l) of the sparse hydroacoustic system at time l, the estimated weight vector h(l) at time l, and The input signal x(l) is obtained when
[0052] The output value of the i-th channel at time l is obtained based on the Lawson norm ;
[0053] According to the gradient descent theory, the estimated weight vector h(l), the estimated error e(l) at time l, and the output value of the i-th channel at time l are used. Get the feature matrix , for the feature matrix The individual items of are operated to obtain the transition matrix G(l), and further combined with the estimated weight vector h(l) to obtain the second transition matrix s(l).
[0054] Among them, in this embodiment:
[0055] Consider a Weight vector of unknown underwater acoustic sparse system ,
[0056] Indicates time When the input signal is The output of the time-sparse hydroacoustic system is: ,in is the environmental noise. The Lawson norm and error of the tap coefficient vector The RLS function constructs the cost function:
[0057]
[0058] in, For the moment The estimated weight vector of is the estimation error at time l, which can be expressed as , Expressed as the traditional RLS cost function, is the forgetting factor, which controls the error weight at different times. is a very small constant. The Lawson norm is expressed as:
[0059]
[0060] in, is a small constant. When equal to 0 or 1, the Lawson norm is approximately Norm and Norm.
[0061] According to the gradient descent theory, there is a feature matrix The formula expansion shows:
[0062]
[0063] Calculate each term of the above formula separately, and we have:
[0064]
[0065] The second term is expressed as:
[0066]
[0067] The above formula can be rewritten as:
[0068]
[0069] Here we define:
[0070]
[0071] in
[0072]
[0073] The initial value of the filter Set to a K-dimensional 0 vector.
[0074] Furthermore, the step 2 specifically includes:
[0075] Set the feature matrix The initial value is , then the gain matrix To solve;
[0076] Order the feature matrix The initial value of is 0, and the solution of the tap coefficient of the adaptive filter is solved;
[0077] Define the weighted complex autocorrelation matrix and solve the gain vector according to the matrix inversion lemma .
[0078] In this embodiment, the characteristic matrix is set The initial value is , then the gain matrix To solve.
[0079] By calculation To solve the tap coefficients of this adaptive filter, we have:
[0080]
[0081] From the above formula, we can see that The acquisition of requires the definition of a weighted complex autocorrelation matrix :
[0082]
[0083] Computed by the matrix inversion lemma ,get:
[0084]
[0085] Its initial value , is a small positive real number, is the gain vector, expressed as:
[0086] .
[0087] Furthermore, the step 3 specifically includes:
[0088] According to the gain vector , feature matrix , the estimated error e(l) at time l, and the transition matrix s(l) are used to obtain the iterative formula of the Lawson-RLS algorithm.
[0089] In this embodiment, by rearranging the above formulas, the iterative formula of the Lawson-RLS algorithm can be easily obtained by calculation:
[0090] .
[0091] Furthermore, the step 4 specifically includes:
[0092] The iteration number condition is set to the length of the received signal, using the prior information of the transmitted signal and the received signal.
[0093] Figure 2 It is an algorithm based on Lawson norm and RLS error function and traditional RLS and -RLS MSD curve, the background noise is a pulse noise with a generalized signal-to-noise ratio of 0dB. This method is significantly better than the traditional RLS method and the optimized -RLS method.
[0094] Figure 3 This is the MSD curve comparing the algorithm based on Lawson norm and RLS error function with several other existing norm-based channel estimation algorithms. When the p value is 0, this method is significantly better than other algorithms and has good convergence speed and steady-state error.
[0095] Figure 4 It is the polar ice cracking noise collected by the 11th Arctic scientific expedition.
[0096] Figure 5 The channel estimation results based on the Lawson norm and RLS error function after adding polar ice crack noise show that the proposed method can accurately estimate the amplitude and delay of underwater acoustic multipath, and the channel reconstruction capability reaches 99.73%.
[0097] Example 2
[0098] According to a second embodiment of the present invention, the present invention claims protection for a robust estimation device for underwater acoustic channels resistant to ice field noise, comprising:
[0099] a memory for storing non-transitory computer-readable instructions; and
[0100] A processor is used to run the computer-readable instructions so that the processor implements the robust estimation method of an underwater acoustic channel resistant to ice sheet noise when executing the instructions.
[0101] Those skilled in the art will appreciate that the contents disclosed in this disclosure may be subject to various modifications and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.
[0102] Flowcharts are used in this disclosure to illustrate the steps of the method according to the embodiments of the present disclosure. It should be understood that the preceding or following steps are not necessarily performed precisely in order. On the contrary, various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes.
[0103] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of a software functional module. The present disclosure is not limited to any particular form of combination of hardware and software.
[0104] Unless otherwise defined, all terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present disclosure belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or highly formal sense, unless explicitly defined as such herein.
[0105] The above is an explanation of the present disclosure and should not be considered as a limitation thereof. Although several exemplary embodiments of the present disclosure are described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present disclosure. Therefore, all such modifications are intended to be included within the scope of the present disclosure as defined in the claims. It should be understood that the above is an explanation of the present disclosure and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present disclosure is defined by the claims and their equivalents.
[0106] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0107] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.
Claims
1. A robust estimation method for underwater acoustic channels resistant to ice sheet noise. It is characterized in that include: S1: constructing a cost function of the Lawson-RLS algorithm, obtaining an estimated weight vector and an estimated error according to the cost function, constructing a characteristic matrix based on the Lawson norm, and presetting parameters of the characteristic matrix through transmitting and receiving signals; S1 specifically includes the following steps: S10. Using a multi-dimensional weight vector of an unknown hydroacoustic sparse system; S11. Based on the input signal and the ambient noise at time l, and in combination with the weight vector, the output of the sparse hydroacoustic system at time l is obtained; S12. Construct a cost function using the Lawson norm of the tap coefficient vector and the RLS function of the error e(l); wherein the cost function is expressed as: Among them, λ∈(0,1) is the forgetting factor; M is the length of the estimated weight vector h(l) at time l; i is the number of iterations; f(e(i)) = e(i)e(i) * is the RLS cost function; η>0; h(l) is the estimated weight vector at time l; ‖h(l)‖ L is the Lawson norm, Where β>0, p is a parameter. When p is equal to 0 or 1, the Lawson norm is approximately l 0 Norm and l 1 norm; S13. Obtaining the estimated weight vector h(l) at time l and the estimated error e(l) at time l according to the cost function; S14. Obtaining output values of different multipath channels at time l based on the Lawson norm; S15. According to the gradient descent theory, each single item of the feature matrix is operated to obtain a transition matrix s(l), s(l)=G(l)h(l), where G(l) is a differential operator matrix; S2: Setting the initial value of the characteristic matrix, solving the gain matrix, and obtaining the gain vector; S2 specifically includes the following steps: S20. Setting the initial value Φ(0) of the characteristic matrix Φ ―1 =f ―1 I M , where f is a positive real number, I M is the M-order unit matrix; solve the gain matrix k; S21. The initial value of the characteristic matrix Φ is set to 0, and the solution of the tap coefficients of the adaptive filter is solved; S22. Define a weighted complex autocorrelation matrix, and solve the gain vector k(l) according to the matrix inversion lemma; S3: Obtain the iterative formula h(l) of the Lawson-RLS algorithm according to the gain vector k(l), the characteristic matrix Φ, the estimated error e(l) at time l, and the transition matrix s(l). S4: Update the characteristic matrix, and continue to repeat the above steps 1-3 until step 3 meets the iteration number condition, and obtain the estimated values of the channel delay and amplitude.
2. The method for robustly estimating underwater acoustic channels resistant to ice sheet noise according to claim 1, It is characterized in that The step 4 specifically includes: The iteration number condition is set to the length of the received signal, using the prior information of the transmitted signal and the received signal.
3. A robust estimation device for underwater acoustic channels resistant to ice sheet noise. include: a memory for storing non-transitory computer-readable instructions; as well as A processor is used to run the computer-readable instructions so that when the processor executes, a robust estimation method for an underwater acoustic channel resistant to ice sheet noise as described in any one of claims 1 to 2 is implemented.
