Low-complexity frequency domain sparse adaptive line spectrum enhancement method and system

Abstract

The invention provides a low-complexity frequency domain sparse adaptive line spectrum enhancement method and system. The method comprises the following steps: S1, reading a radiation noise signal generated by an underwater acoustic target; s2, dividing the decorrelation signal into snapshots, and taking the snapshots as time domain input of an adaptive filter; s3, obtaining a frequency domain snapshot; s4, taking the frequency domain snapshot as the frequency domain input of the adaptive filter; s5, calculating to obtain a filtered output signal and a corresponding estimation error; s6, constructing a cost function and introducing a norm of a frequency domain tap weight vector as a regularization item; step S7, obtaining an updated frequency domain tap weight vector; step S8, obtaining time domain input of the next moment; and S9, circulating the steps S3 to S8 to obtain an enhanced radiation noise signal. According to the method, the time domain input is converted into the frequency domain by adopting the sliding discrete Fourier transform, the calculation complexity is reduced from O (L) to O (Llog2L), the calculation complexity of the ALE is greatly reduced, and a foundation is laid for engineering application.

Description

Technical Field

[0001] The present invention relates to the field of sonar technology in underwater acoustic engineering, and in particular to a low-complexity frequency domain sparse adaptive line spectrum enhancement method and system. Background Art

[0002] The spectrum of the radiated noise of the underwater acoustic target shows the characteristics of the superposition of the continuous spectrum and the line spectrum. Among them, the sources of the line spectrum mainly include the reciprocating operation of the mechanical parts of the underwater acoustic target, the periodic water impact of the propeller, and the resonance phenomenon of the target itself. Since the underwater acoustic target generally has a large inertia and the working conditions are relatively stable, its line spectrum often has a high degree of stability. This stability makes the line spectrum an extremely critical basis for underwater acoustic target detection and identification.

[0003] However, with the continuous progress and widespread application of modern vibration reduction and noise reduction technology, the noise level radiated by underwater acoustic targets has shown a decreasing trend year by year. In this context, line spectrum is easily masked by strong interference and difficult to stand out. Therefore, breaking through the line spectrum enhancement technology under low signal-to-noise ratio conditions has extremely important practical significance and application value for improving the detection and recognition capabilities of underwater acoustic targets. In the current field of underwater acoustic signal processing, adaptive line spectrum enhancer (ALE) is a commonly used line spectrum enhancement method.

[0004] Through the search of patent documents, it was found that the patent document with application number 202311269192.1 discloses a frequency-domain sparse adaptive line spectrum enhancement method. This method is based on the sparse characteristics of the ALE frequency response, and uses a sparse adaptive algorithm to iteratively update the frequency-domain tap weight vector of the ALE. In this way, it makes it closer to the Wiener solution of the ALE, thereby effectively improving the processing gain and broadening the application range of the signal-to-noise ratio.

[0005] Despite this, ALE faces severe challenges in practical engineering applications. The root cause is that the working mechanism adopted by ALE is a point-by-point sliding processing mode. This means that the input vector needs to be updated in each time step. That is, the inner product operation is first performed on the current input vector and the tap weight vector of ALE to generate the output signal of ALE, and then the error is calculated based on the output signal and the tap weight vector is updated accordingly to optimize the performance of ALE. After completing the above operations, slide another sampling point to obtain the input vector at the next moment, and repeat the above series of tedious operations. It is precisely because of these complex and frequent calculation processes that the computational complexity of ALE remains high, which greatly limits its wide application and promotion in practical engineering.

[0006] In summary, in response to the above-mentioned problems of the prior art, studying a low-complexity frequency-domain sparse adaptive line spectrum enhancement method and system to provide a more efficient and practical line spectrum enhancement solution for underwater acoustic signal processing has become a key task that needs to be solved urgently. Summary of the invention

[0007] In view of the defects in the prior art, an object of the present invention is to provide a low-complexity frequency-domain sparse adaptive line spectrum enhancement method and system.

[0008] A low-complexity frequency-domain sparse adaptive line spectrum enhancement method provided by the present invention comprises the following steps:

[0009] Step S1, reading the radiation noise signal generated by the hydroacoustic target as a reference signal for the adaptive filter;

[0010] Step S2, performing a delay decoupling operation on the radiation noise signal to obtain a decorrelated signal, and then dividing the decorrelated signal into snapshots according to the length of the tap weight vector of the adaptive filter as a time domain input of the adaptive filter;

[0011] Step S3, using a sliding discrete Fourier transform to transform the time domain input into the frequency domain to obtain a frequency domain snapshot;

[0012] Step S4, using the frequency domain snapshot as the frequency domain input of the adaptive filter;

[0013] Step S5, calculating a filtered output signal and a corresponding estimation error based on a reference signal and a frequency domain input of the adaptive filter;

[0014] Step S6, constructing a cost function based on the maximum correlation entropy criterion according to the estimation error, wherein the cost function introduces the norm of the frequency domain tap weight vector as a regularization term;

[0015] Step S7, calculating the gradient of the cost function, deriving the updating formula of the frequency domain tap weight vector of ALE, and obtaining the updated frequency domain tap weight vector;

[0016] Step S8, sliding the decorrelation signal by one sampling point to obtain the time domain input at the next moment;

[0017] Step S9, looping through steps S3 to S8 until all radiation noise signals are processed to obtain enhanced radiation noise signals.

[0018] Preferably, step S2 includes the following sub-steps:

[0019] Step S2.1, performing a delay operation on the radiation noise signal s(n) to obtain a decorrelated signal s(n-Δ);

[0020] Step S2.2, according to the length of the tap weight vector of the adaptive filter, the decorrelated signal s(n-Δ) is divided into snapshots as the time domain input of the adaptive filter at time n, and the calculation formula is:

[0021] s(n-Δ)=[s(n-Δ),…,s(n-Δ-L+1)] T (1)

[0022] Wherein, L is the length of the tap weight vector of the adaptive filter, Δ is the delay time, and T represents transposition.

[0023] Preferably, in step S2.2, L=1000, Δ=1.

[0024] Preferably, in step S3, the formula for the frequency domain input of the adaptive filter at time n is:

[0025]

[0026] Among them, s F (n-Δ) is the discrete Fourier transform of the time domain input, are the frequency components of discrete Fourier transform, i=0,…,L-1;

[0027] The derivation formula is:

[0028]

[0029] Here, j represents the imaginary unit, and m and p represent the summation index.

[0030] Based on the derived formula, the frequency domain input s of the adaptive filter at time n F (n-Δ) is the frequency domain input s of the adaptive filter at time n-1 F (n-1-Δ) represents.

[0031] Preferably, in step S5, for each time instant n, the calculation formula of the filtered output signal is as follows:

[0032]

[0033] in, is the frequency domain tap weight vector of the adaptive filter at time n, with an initial value of 0;

[0034] The corresponding estimation error is calculated as:

[0035] e(n)=s(n)-y(n) (5)

[0036] Where s(n) is the input signal, (·) H represents the conjugate transpose.

[0037] Preferably, in step S6, the cost function is calculated as follows:

[0038]

[0039] Among them, σ represents the kernel width of the maximum correlation entropy criterion; ||·||1 represents the L1 norm; ρ>0 represents the balance factor, which is used to balance the influence of the regularization term; exp represents the exponential function.

[0040] Preferably, in step S6, σ=1; ρ=0.01.

[0041] Preferably, in step S7, the updating formula of the frequency domain tap weight vector is as follows:

[0042]

[0043] Where μ represents the update step size of the frequency domain tap weight vector; (·) * represents the conjugate operation; sgn(·) represents the sign function.

[0044] Preferably, in step S7, μ=10 -5 .

[0045] The present invention also provides a low-complexity frequency-domain sparse adaptive line spectrum enhancement system, comprising:

[0046] Module M1, reads the radiation noise signal generated by the hydroacoustic target as a reference signal for the adaptive filter;

[0047] Module M2 performs a delay decoupling operation on the radiation noise signal to obtain a decorrelated signal, and then divides the decorrelated signal into snapshots according to the length of the tap weight vector of the adaptive filter as the time domain input of the adaptive filter;

[0048] Module M3, uses sliding discrete Fourier transform to transform the time domain input to the frequency domain to obtain the frequency domain snapshot;

[0049] Module M4, taking the frequency domain snapshot as the frequency domain input of the adaptive filter;

[0050] Module M5, based on the reference signal and frequency domain input of the adaptive filter, calculates the filtered output signal and the corresponding estimation error;

[0051] Module M6, constructs a cost function based on the maximum correlation entropy criterion according to the estimation error, and the cost function introduces the norm of the frequency domain tap weight vector as a regularization term;

[0052] Module M7, calculates the gradient of the cost function, derives the update formula of the frequency domain tap weight vector of ALE, and obtains the updated frequency domain tap weight vector;

[0053] Module M8 slides the decorrelated signal by one sampling point to obtain the time domain input at the next moment;

[0054] Module M9 cyclically executes module M3 , module M4 , module M5 , module M6 , module M7 and module M8 until all radiation noise signals are processed to obtain enhanced radiation noise signals.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. The present invention uses sliding discrete Fourier transform to convert time domain input to frequency domain, reducing the computational complexity from O(L) to O(Llog2L), greatly reducing the computational complexity of ALE, thus laying a foundation for engineering applications.

[0057] 2. The present invention designs the cost function of ALE based on the maximum correlation entropy criterion, so that ALE still has good applicability in non-Gaussian noise environment, effectively expanding its application scenarios.

[0058] 3. The present invention introduces a regularization term into the cost function of ALE to encourage the sparsity of the frequency domain tap weight vector, so that the tap weight vector of ALE is closer to the Wiener solution, thereby improving the processing gain of ALE. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0060] Figure 1 A flowchart of a low-complexity frequency-domain sparse adaptive line spectrum enhancement method in an embodiment of the present invention;

[0061] Figure 2 A time domain waveform diagram of non-Gaussian noise in an embodiment of the present invention;

[0062] Figure 3 LOFAR spectrum of the input signal in the embodiment of the present invention;

[0063] Figure 4 The frequency domain tap weight vector of the conventional ALE in the embodiment of the present invention;

[0064] Figure 5 The frequency domain tap weight vector of the method of the present invention in the embodiment of the present invention;

[0065] Figure 6 LOFAR spectrum of conventional ALE output signal in an embodiment of the present invention;

[0066] Figure 7 The LOFAR spectrum of the output signal of the method of the present invention in the embodiment of the present invention. DETAILED DESCRIPTION

[0067] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0068] The technical problem to be solved by the present invention is to provide a low-complexity frequency-domain sparse adaptive line spectrum enhancement method to solve the problem that the frequency-domain sparse adaptive line spectrum enhancement method has high complexity and is difficult to apply in engineering.

[0069] Embodiment 1:

[0070] Figure 1 A flowchart of a low-complexity frequency-domain sparse adaptive line spectrum enhancement method in an embodiment of the present invention.

[0071] like Figure 1 As shown, this embodiment provides a low-complexity frequency domain sparse adaptive line spectrum enhancement method, comprising the following steps:

[0072] Step S1, reading the radiation noise signal s(n) generated by the hydroacoustic target as a reference signal for the adaptive filter.

[0073] In this embodiment, the input signal is x(n)=Acos(2πfn / f s )+noise(n), line spectrum frequency f=100Hz, sampling rate f s =2000Hz, signal length 20s, signal-to-noise ratio -27dB, and input noise is non-Gaussian noise that obeys mixed Gaussian distribution.

[0074] Figure 2 A time domain waveform diagram of non-Gaussian noise in an embodiment of the present invention; Figure 3 LOFAR spectrum of the input signal in the embodiment of the present invention.

[0075] Reference for the time domain waveform of non-Gaussian noise Figure 2 As shown in the figure, compared with Gaussian noise, non-Gaussian noise has a "heavy tail" characteristic, which leads to a larger dynamic range of non-Gaussian noise, which is not conducive to the detection of line spectrum; the LOFAR spectrum of the input signal is referenced Figure 3 As shown, it can be seen that the signal-to-noise ratio of the input signal is extremely low and the line spectrum is completely submerged by the noise.

[0076] Step S2, performing a delay decoupling operation on the radiation noise signal s(n) to obtain a decorrelated signal, and then dividing the decorrelated signal into snapshots according to the length of the tap weight vector of the adaptive filter as the time domain input of the adaptive filter.

[0077] Specifically, step S2 includes the following sub-steps:

[0078] Step S2.1, performing a delay operation on the radiation noise signal s(n) to obtain a decorrelated signal s(n-Δ).

[0079] Step S2.2, according to the length of the adaptive filter tap weight vector, the decorrelated signal s(n-Δ) is divided into snapshots as the time domain input of the adaptive filter at time n, and the calculation formula is:

[0080] s(n-Δ)=[s(n-Δ),…,s(n-Δ-L+1)] T (1)

[0081] Wherein, L is the length of the adaptive filter tap weight vector, Δ is the delay time, and T represents transposition.

[0082] In this embodiment, L=1000, Δ=1.

[0083] Step S3, using sliding discrete Fourier transform to transform the time domain input into the frequency domain to obtain a frequency domain snapshot.

[0084] Specifically, the formula for the frequency domain input of the adaptive filter at time n is:

[0085]

[0086] Among them, s F (n-Δ) is the discrete Fourier transform of the time domain input, are the frequency components of discrete Fourier transform, i=0,…,L-1.

[0087] Specifically,

[0088]

[0089] Here, j represents the imaginary unit, and m and p represent the summation index.

[0090] It can be seen that the frequency domain input s of the adaptive filter at time n is F (n-Δ) is the frequency domain input s of the adaptive filter at time n-1 F (n-1-Δ) represents.

[0091] Compared with the discrete Fourier transform and the fast discrete Fourier transform, the sliding discrete Fourier transform has significant advantages in transforming the time domain input to the frequency domain. First, the sliding discrete Fourier transform and the discrete Fourier transform have the same effect and will not cause the loss of frequency domain information. Second, the sliding discrete Fourier transform only requires one complex multiplication operation and two real addition operations to transform the time domain input to the frequency domain, with a computational complexity of O(L), while the computational complexity of the discrete Fourier transform and the fast discrete Fourier transform is O(L) respectively. 2 ) and Ο(Llog2L). It can be seen that the complexity of the sliding discrete Fourier transform is low, and its advantage is more obvious when the adaptive filter tap weight vector L is large.

[0092] Step S4: using the frequency domain snapshot as the frequency domain input of the adaptive filter.

[0093] Step S5, based on the reference signal and the frequency domain input of the adaptive filter, a filtered output signal and a corresponding estimation error are calculated.

[0094] Specifically, for each time instant n, the calculation formula of the filtered output signal is as follows:

[0095]

[0096] in, is the frequency domain tap weight vector of the adaptive filter at time n, with an initial value of 0.

[0097] The corresponding estimation error is calculated as:

[0098] e(n)=s(n)-y(n) (5)

[0099] Where s(n) is the input signal, (·) H represents the conjugate transpose.

[0100] Step S6: construct a cost function based on the maximum correlation entropy criterion according to the estimation error. The cost function introduces the L1 norm of the frequency domain tap weight vector as a regularization term to encourage the sparsity of the frequency domain tap weight vector.

[0101] Specifically, the calculation formula of the cost function is:

[0102]

[0103] Among them, σ represents the kernel width of the maximum correlation entropy criterion; ||·||1 represents the L1 norm; ρ>0 represents the balance factor, which is used to balance the influence of the regularization term; exp represents the exponential function.

[0104] In this embodiment, σ=1; ρ=0.01.

[0105] Step S7, calculate the gradient of the cost function, derive the update formula of the frequency domain tap weight vector of ALE, and obtain the updated frequency domain tap weight vector.

[0106] Specifically, the updating formula of the frequency domain tap weight vector is as follows:

[0107]

[0108] Where μ represents the update step size of the frequency domain tap weight vector; (·) * represents the conjugate operation; sgn(·) represents the sign function.

[0109] In this embodiment, μ=10 -5 .

[0110] Step S8, sliding the decorrelation signal by one sampling point to obtain the time domain input at the next moment;

[0111] Step S9, looping through steps S3, S4, S5, S6, S7 and S8 until all radiation noise signals are processed to obtain enhanced radiation noise signals.

[0112] Figure 4 The frequency domain tap weight vector of the conventional ALE in the embodiment of the present invention; Figure 5 The frequency domain tap weight vector of the method of the present invention in the embodiment of the present invention; Figure 6 LOFAR spectrum of conventional ALE output signal in an embodiment of the present invention; Figure 7 The LOFAR spectrum of the output signal of the method of the present invention in the embodiment of the present invention.

[0113] like Figure 4 , 5 As shown, due to the influence of the regularization term, the frequency domain tap weight vector of the method of the present invention is closer to the Wiener solution, which can better suppress broadband noise and achieve higher processing gain.

[0114] like Figure 6 , 7 As shown, the line spectrum enhanced by the method of the present invention is clearly visible and is significantly better than conventional ALE.

[0115] Embodiment 2:

[0116] The present invention also provides a low-complexity frequency-domain sparse adaptive line spectrum enhancement system. The low-complexity frequency-domain sparse adaptive line spectrum enhancement system can be realized by executing the process steps of a low-complexity frequency-domain sparse adaptive line spectrum enhancement method. That is, those skilled in the art can understand the low-complexity frequency-domain sparse adaptive line spectrum enhancement method as a preferred implementation of the low-complexity frequency-domain sparse adaptive line spectrum enhancement system.

[0117] The frequency domain sparse adaptive line spectrum enhancement system includes:

[0118] Module M1, reads the radiation noise signal generated by the hydroacoustic target as a reference signal for the adaptive filter;

[0119] Module M2 performs a delay decoupling operation on the radiation noise signal to obtain a decorrelated signal, and then divides the decorrelated signal into snapshots according to the length of the tap weight vector of the adaptive filter as the time domain input of the adaptive filter;

[0120] Module M3, uses sliding discrete Fourier transform to transform the time domain input to the frequency domain to obtain the frequency domain snapshot;

[0121] Module M4, taking the frequency domain snapshot as the frequency domain input of the adaptive filter;

[0122] Module M5, based on the reference signal and frequency domain input of the adaptive filter, calculates the filtered output signal and the corresponding estimation error;

[0123] Module M6, constructs a cost function based on the maximum correlation entropy criterion according to the estimation error, and the cost function introduces the norm of the frequency domain tap weight vector as a regularization term;

[0124] Module M7, calculates the gradient of the cost function, derives the update formula of the frequency domain tap weight vector of ALE, and obtains the updated frequency domain tap weight vector;

[0125] Module M8 slides the decorrelated signal by one sampling point to obtain the time domain input at the next moment;

[0126] Module M9 cyclically executes module M3 , module M4 , module M5 , module M6 , module M7 and module M8 until all radiation noise signals are processed to obtain enhanced radiation noise signals.

[0127] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0128] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A low-complexity frequency-domain sparse adaptive line spectrum enhancement method, characterized in that: The following steps are involved: Step S1, reading the radiation noise signal generated by the hydroacoustic target as a reference signal for the adaptive filter; Step S2, performing a delay decoupling operation on the radiation noise signal to obtain a decorrelation signal, and then dividing the decorrelation signal into snapshots according to the length of the tap weight vector of the adaptive filter as a time domain input of the adaptive filter; Step S3, transforming the time domain input into the frequency domain using a sliding discrete Fourier transform to obtain a frequency domain snapshot; Step S4, using the frequency domain snapshot as a frequency domain input of an adaptive filter; Step S5, calculating a filtered output signal and a corresponding estimation error based on a reference signal and a frequency domain input of the adaptive filter; Step S6, constructing a cost function based on the maximum correlation entropy criterion according to the estimation error, wherein the cost function introduces the norm of the frequency domain tap weight vector as a regularization term; Step S7, calculating the gradient of the cost function, deriving an update formula for the frequency domain tap weight vector of ALE, and obtaining an updated frequency domain tap weight vector; Step S8, sliding the decorrelation signal by one sampling point to obtain a time domain input at the next moment; Step S9, looping through steps S3, S4, S5, S6, S7 and S8 until all radiation noise signals are processed to obtain enhanced radiation noise signals.

2. A low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 1, characterized in that: The step S2 includes the following sub-steps: Step S2.1, performing a delay operation on the radiation noise signal s(n) to obtain a decorrelated signal s(n-Δ); Step S2.2, according to the length of the tap weight vector of the adaptive filter, the decorrelated signal s(n-Δ) is divided into snapshots as the time domain input of the adaptive filter at time n, and the calculation formula is: s(n-Δ)=[s(n-Δ),…,s(n-Δ-L+1)] T (1) Wherein, L is the length of the tap weight vector of the adaptive filter, Δ is the delay time, and T represents transposition.

3. A low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 2, characterized in that: In step S2.2, L=1000, Δ=1.

4. The low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 1, characterized in that: In step S3, the formula of the frequency domain input of the adaptive filter at time n is: Among them, s F (n-Δ) is the discrete Fourier transform of the time domain input, are the frequency components of discrete Fourier transform, i=0,…,L-1; The derivation formula is: Among them, j represents the imaginary unit, m and p represent the summation index; Based on the derived formula, the frequency domain input s of the adaptive filter at time n F (n-Δ) is the frequency domain input s of the adaptive filter at time n-1 F (n-1-Δ) represents.

5. The low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 1, characterized in that: In step S5, for each time instant n, the calculation formula of the filtered output signal is as follows: in, is the frequency domain tap weight vector of the adaptive filter at time n, with an initial value of 0; The corresponding estimation error is calculated as: e(n)=s(n)-y(n) (5) Where s(n) is the input signal, (·) H represents the conjugate transpose.

6. The low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 1, characterized in that: In step S6, the calculation formula of the cost function is: Among them, σ represents the kernel width of the maximum correlation entropy criterion; ||·||1 represents the L1 norm; ρ>0 represents the balance factor, which is used to balance the influence of the regularization term; exp represents the exponential function.

7. A low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 6, characterized in that: In the step S6, σ=1; ρ=0.

01.

8. The low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 1, characterized in that: In step S7, the updating formula of the frequency domain tap weight vector is as follows: Where μ represents the update step size of the frequency domain tap weight vector; (·) * represents the conjugate operation; sgn(·) represents the sign function.

9. A low-complexity frequency-domain sparse adaptive line spectrum enhancement method according to claim 8, characterized in that: In step S7, μ=10 -5 .

10. A low-complexity frequency-domain sparse adaptive line spectrum enhancement system, characterized in that: include: Module M1, reads the radiation noise signal generated by the hydroacoustic target as a reference signal for the adaptive filter; Module M2, performing a delay decoupling operation on the radiation noise signal to obtain a decorrelation signal, and then dividing the decorrelation signal into snapshots according to the length of the tap weight vector of the adaptive filter as a time domain input of the adaptive filter; Module M3, transforming the time domain input into the frequency domain using a sliding discrete Fourier transform to obtain a frequency domain snapshot; Module M4, using the frequency domain snapshot as a frequency domain input of an adaptive filter; Module M5, based on the reference signal and frequency domain input of the adaptive filter, calculates the filtered output signal and the corresponding estimation error; Module M6, constructing a cost function based on the maximum correlation entropy criterion according to the estimation error, wherein the cost function introduces the norm of the frequency domain tap weight vector as a regularization term; Module M7, calculates the gradient of the cost function, derives the update formula of the frequency domain tap weight vector of ALE, and obtains the updated frequency domain tap weight vector; Module M8, sliding the decorrelation signal by one sampling point to obtain the time domain input at the next moment; Module M9 cyclically executes module M3 , module M4 , module M5 , module M6 , module M7 and module M8 until all radiation noise signals are processed to obtain enhanced radiation noise signals.

Images