Helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform

Abstract

The application discloses a helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transformation, and comprises the following steps: collecting acoustic signals emitted by a low-altitude helicopter target by using a microphone array and performing pretreatment, wherein the helicopter acoustic signal is a typical harmonic signal; performing frequency focusing processing on the helicopter acoustic signals collected by each microphone by using phase-corrected frequency rearrangement synchronous compression transformation; performing short-time inverse Fourier transformation on the frequency domain results after the phase-corrected frequency rearrangement synchronous compression transformation to obtain time domain multi-channel helicopter signals after the phase-corrected frequency rearrangement synchronous compression transformation; and performing adaptive line spectrum enhancement on the time domain multi-channel helicopter signals after the phase-corrected frequency rearrangement synchronous compression transformation to realize enhancement of the helicopter acoustic signal. The helicopter acoustic signal has improved signal-to-noise ratio under low signal-to-noise ratio by using the method combining adaptive line spectrum enhancement and synchronous compression transformation.

Description

Technical Field

[0001] This invention belongs to the field of acoustic signal enhancement and sound source localization technology, and in particular, it is a method for enhancing helicopter acoustic signals by combining adaptive line spectrum enhancement and synchronous compression transformation. Background Technology

[0002] Helicopters are a typical dual-use product, widely used in transportation, patrol, tourism, disaster relief, and localized strikes. However, they also pose problems such as unauthorized helicopter flights, cross-border smuggling of illegal drugs, illegal immigration, and other illegal activities. Therefore, the detection, identification, location, and tracking of helicopters are receiving increasing attention. However, the low signal-to-noise ratio of long-range helicopter acoustic signals captured by microphone arrays makes helicopter target detection extremely difficult. The key lies in reducing environmental noise interference with the desired helicopter acoustic signal and enhancing the helicopter's audio information to improve the signal-to-noise ratio.

[0003] Currently, the main signal enhancement algorithms for helicopter acoustic signals include: 1) wavelet transform-based methods, the performance of which is highly dependent on parameter selection, and there is currently no good standard for parameter selection; 2) methods based on higher-order statistics, but these methods require a large number of data samples, have high computational complexity, and are poorly effective for signals with low signal-to-noise ratios; 3) beamforming, which requires knowledge of the target's azimuth, and the enhancement performance is limited by the number of array elements and the array aperture; 4) synchronous compression transform, which is mainly used in fields such as fault diagnosis of rotating machinery systems and ultrasonic guided wave signal analysis, but has not been applied to the field of helicopter acoustic signal enhancement.

[0004] Therefore, a method is needed to reduce the impact of environmental noise and other interference on helicopter acoustic signals, so that acoustic detection equipment can extract target helicopter acoustic signals with high signal-to-noise ratio in complex background noise. Summary of the Invention

[0005] The purpose of this invention is to address the problems existing in the prior art by providing a helicopter acoustic signal enhancement method that combines adaptive line spectrum enhancement and synchronous compression transformation, thereby improving the signal-to-noise ratio of helicopter acoustic signals and laying the foundation for improving the direction-finding accuracy of helicopter targets.

[0006] The technical solution to achieve the objective of this invention is: a helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform, the method comprising the following steps:

[0007] Step 1: Acquire multi-channel helicopter acoustic signals using a microphone array and perform preprocessing.

[0008] Step 2: Perform phase-corrected frequency rearrangement synchronization compression transformation on the preprocessed helicopter acoustic signal;

[0009] Step 3: Perform an inverse short-time Fourier transform on the frequency domain result after the phase-corrected frequency rearrangement synchronous compression transform.

[0010] Step 4: Perform adaptive line spectrum enhancement on the multi-channel signals after inverse short-time Fourier transform to obtain the enhanced time-domain multi-channel helicopter acoustic signals.

[0011] Further, the preprocessing in step 1 includes:

[0012] Step 1-1: Eliminate the DC component from the helicopter's acoustic signal;

[0013] Steps 1-2 involve frame segmentation.

[0014] Furthermore, step 2, which involves performing phase-corrected frequency rearrangement synchronization compression transformation on the preprocessed helicopter acoustic signal, specifically includes:

[0015] Step 2-1: Perform a windowed short-time Fourier transform on the helicopter acoustic signals of each channel to obtain:

[0016]

[0017]

[0018] Where, x i [m] represents the helicopter sound signal collected by the i-th microphone at sampling time m, i.e., the helicopter sound signal of the i-th channel at sampling time m, i = 1, 2, ..., M, where M is the number of microphones in the microphone array, g represents the Gaussian window function, g′ represents the first derivative of the Gaussian window function, * represents the conjugate, n represents the sampling time, n = 0, 1, ..., N-1, where N is the number of sampling points in one frame of the helicopter sound signal, and k represents the frequency, k = 0, 1, ..., f s / 2,f s For the sampling rate, g * [mn] represents the conjugate of the Gaussian window values ​​at sampling time mn, (g′[mn]). * This represents the result of taking the conjugate of the first derivative of the Gaussian window at sampling time mn. Let represent the time-frequency representation of the helicopter acoustic signal of the i-th channel at the sampling time n and frequency k, after Fourier transform using a Gaussian window as the window function, at the instantaneous frequency point [n,k]. The time-frequency representation of the helicopter acoustic signal of the i-th channel at sampling time n and frequency k after Fourier transform using the first derivative of the Gaussian window as the window function;

[0019] Step 2-2: For each time-frequency point of the helicopter acoustic signal in each channel, calculate its instantaneous frequency operator. This instantaneous frequency operator indicates the position of the energy centroid corresponding to each time-frequency point on the time-frequency plane.

[0020]

[0021] Where the control threshold γ is a positive number, and Im[g] represents taking the imaginary part; The frequency of the energy centroid at the sampling time n and frequency k after the Fourier transform of the helicopter acoustic signal of the i-th channel using a Gaussian window as the window function is represented.

[0022] Steps 2-3, at sampling time n, represent the time-frequency of each time-frequency point as shown in equation (4), perform phase correction, and then rearrange it to its energy centroid position. Specifically, this includes:

[0023] The time-frequency representation of the time-frequency point [n,k1] is as follows: The frequency of the energy centroid at that time point can be calculated using equation (3). Then the time-frequency representation of its energy centroid location [n, k0] is as follows: in A1[n,k1] and A0[n,k0] are the amplitudes of time-frequency points [n,k1] and [n,k0], respectively; 2πk1τ1 and 2πk0τ0 are the phases of time-frequency points [n,k1] and [n,k0], respectively; and τ1 and τ0 are the time delays of time-frequency points [n,k1] and [n,k0], respectively.

[0024] The following correction is made to τ1, so that the phase of the time-frequency point [n,k1] becomes 2πk1τ:

[0025]

[0026] in, angle[g] represents taking the phase; after time-frequency rearrangement, the time-frequency representation of the time-frequency point [n,k1] becomes The time-frequency representation of its centroid location [n,k0] becomes...

[0027] Furthermore, the control threshold γ is set to 2.22 × 10⁻⁶. -16 .

[0028] Furthermore, the frequency domain result after the phase-corrected frequency rearrangement synchronous compression transform, as described in step 3, undergoes an inverse short-time Fourier transform, specifically as follows:

[0029]

[0030] in, The time-frequency representation of sampling time n and frequency k after phase correction, frequency rearrangement, synchronous compression transformation, y FSSTi [n] represents the value of the helicopter acoustic signal of the i-th channel at sampling time n after phase-corrected frequency rearrangement synchronous compression transformation.

[0031] Furthermore, step 4 involves adaptive line spectrum enhancement of the multi-channel signals after inverse short-time Fourier transform, resulting in enhanced time-domain multi-channel helicopter acoustic signals, as shown in the formula:

[0032] y FSST+ALEi [n] = w[n] T y FSSTi [n] (6)

[0033] Among them, y FSST+ALEi [n] represents the value of the enhanced time-domain multi-channel helicopter acoustic signal at sampling time n, and the input signal vector y of the adaptive line spectrum enhancer. FSSTi [n] = [y FSSTi [n] y FSSTi [n-1] ... y FSSTi [nL]] T L is the filter length, y FSSTi [n]、y FSSTi [n-1] and y FSSTi [nL] represent the values ​​of the time-domain helicopter acoustic signal of the i-th channel after phase-corrected frequency rearrangement synchronous compression transformation at sampling times n, n-1, and nL, respectively. The filter weight vector is composed of the weights w of each tap of the filter. j [n] is composed of j = 0, 1, ..., L, and the filter weight vector w[n] = [w0[n] w1[n] … w L [n] T .

[0034] Furthermore, the filter weight vector is updated according to the following formula:

[0035]

[0036] Here, the step size factor α is used to control the increment magnitude of the filter tap weight vector, and c is a constant used to avoid ||y|| . FSSTi [n]|| 2 Too small step size If the value is too large, the error signal e[n] = d[n] - y at sampling time n FSSTi [n], d[n] are the expected values ​​of the current input signal, i.e., the time-domain helicopter acoustic signal after phase correction, frequency rearrangement, synchronous compression transformation, at sampling time n.

[0037] Furthermore, the step size factor α has a range of (0, 2).

[0038] Furthermore, the delay y of the input signal FSSTi [nm] is fed into the adaptive filter as signal y FSSTi The expected value of [n] is d[n], where m is the number of delayed sampling points.

[0039] Compared with the prior art, the significant advantages of this invention are:

[0040] 1) This invention enhances helicopter acoustic signals by combining phase-corrected frequency rearrangement synchronous compression transformation and adaptive line spectrum enhancement algorithm. It can extract helicopter acoustic signals with higher signal-to-noise ratio from complex background noise, greatly improving the ratio of helicopter acoustic signal line spectrum energy to surrounding frequency noise energy. It can be used as front-end processing for helicopter target detection, identification, direction finding and tracking.

[0041] 2) The enhanced helicopter acoustic signal of this invention is based on the MUSIC algorithm for DOA estimation, which can obtain higher direction finding accuracy and realize helicopter target direction finding under low signal-to-noise ratio.

[0042] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0043] Figure 1 This is a flowchart of the helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transformation according to the present invention.

[0044] Figure 2 Here is the time spectrum and power spectrum of a helicopter acoustic signal in one embodiment, wherein Figure 2 (a) in the figure is a time-frequency spectrum. Figure 2 (b) in the diagram is the power spectrum.

[0045] Figure 3 This is a structural diagram of an adaptive line spectrum enhancer in one embodiment.

[0046] Figure 4 Here is the time spectrum and power spectrum of the enhanced helicopter acoustic signal according to one embodiment of the present invention, wherein... Figure 4 (a) in the figure is a time-frequency spectrum. Figure 4 (b) in the diagram is the power spectrum.

[0047] Figure 5 This is a comparison chart of the mean absolute estimation error of the helicopter acoustic signal, the helicopter acoustic signal enhanced by frequency rearrangement synchronous compression transformation, the helicopter acoustic signal enhanced by the adaptive line spectrum enhancement algorithm, and the helicopter acoustic signal enhanced by the present invention, based on the MUSIC algorithm for DOA estimation in one embodiment. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] It should be noted that if the embodiments of the present invention involve descriptions such as "first" and "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" and "second" may explicitly or implicitly include at least one of those features. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0050] In one embodiment, combined Figure 1 This paper presents a method for enhancing helicopter acoustic signals by combining adaptive line spectrum enhancement and synchronous compression transform, comprising the following steps:

[0051] S1, using a microphone array to acquire multi-channel helicopter acoustic signals and perform preprocessing, the time-frequency spectrum of the helicopter acoustic signals is as follows. Figure 2 As shown;

[0052] The preprocessing includes:

[0053] S1-1, for helicopter acoustic signals, eliminates the DC component by removing the mean from the time-domain signal;

[0054] S1-2, perform framing processing. Preferably, the frame length is 1024 and the frame shift is 512.

[0055] S2, performs phase-corrected frequency rearrangement synchronization compression transformation on the preprocessed helicopter acoustic signal. This step specifically includes:

[0056] S2-1, performing a windowed short-time Fourier transform on the helicopter acoustic signals of each channel, yields:

[0057]

[0058]

[0059] Where, x i[m] represents the helicopter sound signal collected by the i-th microphone at sampling time m, i.e., the helicopter sound signal of the i-th channel at sampling time m, i = 1, 2, ..., M, where M is the number of microphones in the microphone array, g represents the Gaussian window function, g′ represents the first derivative of the Gaussian window function, * represents the conjugate, n represents the sampling time, n = 0, 1, ..., N-1, where N is the number of sampling points in one frame of the helicopter sound signal, and k represents the frequency, k = 0, 1, ..., f s / 2,f s For the sampling rate, g * [mn] represents the conjugate of the Gaussian window values ​​at sampling time mn, (g′[mn]). * This represents the result of taking the conjugate of the first derivative of the Gaussian window at sampling time mn. Let represent the time-frequency representation of the helicopter acoustic signal of the i-th channel at the sampling time n and frequency k, after Fourier transform using a Gaussian window as the window function, at the instantaneous frequency point [n,k]. The time-frequency representation of the helicopter acoustic signal of the i-th channel at sampling time n and frequency k after Fourier transform using the first derivative of the Gaussian window as the window function;

[0060] S2-2, calculate the instantaneous frequency operator for each time-frequency point of each channel. This operator indicates the position of the energy centroid corresponding to each time-frequency point on the time-frequency plane:

[0061]

[0062] The control threshold γ is a very small positive number, typically 2.22 × 10⁻⁶. -16 Im[g] indicates taking the imaginary part. The frequency of the energy centroid at the sampling time n and frequency k after the Fourier transform of the helicopter acoustic signal of the i-th channel using a Gaussian window as the window function is represented.

[0063] S2-3, at time n, the time-frequency representation of each time-frequency point is rearranged to its energy centroid position after phase correction according to equation (11), specifically including:

[0064] The time-frequency representation of the time-frequency point [n,k1] is as follows: The frequency of the energy centroid at that time point is calculated according to equation (10). Then the time-frequency representation of its energy centroid location [n, k0] is as follows: in A1[n,k1] and A0[n,k0] are the amplitudes of time-frequency points [n,k1] and [n,k0], respectively; 2πk1τ1 and 2πk0τ0 are the phases of time-frequency points [n,k1] and [n,k0], respectively; and τ1 and τ0 are the time delays of time-frequency points [n,k1] and [n,k0], respectively. Then, τ1 is modified as follows, so that the phase of time-frequency point [n,k1] becomes 2πk1τ:

[0065]

[0066] in, angle[g] represents taking the phase. After time-frequency rearrangement, the time-frequency representation of the time-frequency point [n,k1] becomes... The time-frequency representation of its centroid location [n,k0] becomes...

[0067] S3, perform an inverse short-time Fourier transform on the frequency domain result after the phase-corrected frequency rearrangement synchronous compression transform, the formula is:

[0068]

[0069] in, The time-frequency representation of sampling time n and frequency k after phase correction, frequency rearrangement, synchronous compression transformation, y FSSTi [n] represents the value of the helicopter acoustic signal of the i-th channel at sampling time n after phase-corrected frequency rearrangement synchronous compression transformation.

[0070] S4 performs adaptive line spectrum enhancement on the multi-channel signals after the inverse short-time Fourier transform. The structure of the adaptive line spectrum enhancer is as follows: Figure 3 As shown, the enhanced time-domain multi-channel helicopter acoustic signal is obtained, and the formula is:

[0071] y FSST+ALEi [n] = w[n] T y FSSTi [n] (13)

[0072] Among them, y FSST+ALEi [n] represents the value of the enhanced time-domain multi-channel helicopter acoustic signal at sampling time n, and the input signal vector y of the adaptive line spectrum enhancer. FSSTi [n] = [y FSSTi [n] y FSSTi [n-1] … y FSSTi [nL]] T L is the filter length, y FSSTi [n]、y FSSTi [n-1] and y FSSTi[nL] represent the values ​​of the time-domain helicopter acoustic signal of the i-th channel after phase-corrected frequency rearrangement synchronous compression transformation at sampling times n, n-1, and nL, respectively. The filter weight vector is composed of the weights w of each tap of the filter. j [n] is composed of j = 0, 1, ..., L, and the filter weight vector w[n] = [w0[n] w1[n] ... w L [n] T It can be updated according to the following formula:

[0073]

[0074] Here, the step size factor α controls the increment magnitude of the filter tap weight vector, with a value range of (0,2) to ensure algorithm convergence, and c is a very small constant to avoid ||y|| > 0. FSSTi [n]|| 2 Too small step size If the value is too large, the error signal e[n] = d[n] - y at sampling time n FSSTi [n], d[n] are the expected values ​​of the current input signal, i.e., the time-domain helicopter acoustic signal after phase correction, frequency rearrangement, synchronous compression transformation, at sampling time n. Typically, the delay y of the input signal is... FSSTi [nm] is fed into the adaptive filter as signal y FSSTi The expected signal d[n] of [n], where m is the number of delayed sampling points.

[0075] In one embodiment, a helicopter acoustic signal enhancement system combining adaptive line spectrum enhancement and synchronous compression transform is provided, the system comprising:

[0076] The first module is used to acquire multi-channel helicopter acoustic signals using a microphone array and to perform preprocessing.

[0077] The second module is used to perform phase-corrected frequency rearrangement synchronous compression transformation on the preprocessed helicopter acoustic signal.

[0078] The third module is used to perform inverse short-time Fourier transform on the frequency domain result after phase-corrected frequency rearrangement synchronous compression transform.

[0079] The fourth module is used to perform adaptive line spectrum enhancement on the multi-channel signals after inverse short-time Fourier transform, so as to obtain the enhanced time-domain multi-channel helicopter acoustic signals.

[0080] Specific limitations regarding the helicopter acoustic signal enhancement system combining adaptive line spectrum enhancement and synchronous compression transform can be found in the limitations of the helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform mentioned above, and will not be repeated here. Each module in the aforementioned helicopter acoustic signal enhancement system combining adaptive line spectrum enhancement and synchronous compression transform can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the corresponding operations of each module.

[0081] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements:

[0082] Step 1: Acquire multi-channel helicopter acoustic signals using a microphone array and perform preprocessing;

[0083] Step 2: Perform phase-corrected frequency rearrangement synchronization compression transformation on the preprocessed helicopter acoustic signal;

[0084] Step 3: Perform an inverse short-time Fourier transform on the frequency domain result after the phase-corrected frequency rearrangement synchronous compression transform.

[0085] Step 4: Perform adaptive line spectrum enhancement on the multi-channel signals after inverse short-time Fourier transform to obtain the enhanced time-domain multi-channel helicopter acoustic signals.

[0086] For specific limitations on each step, please refer to the limitations of the helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transformation mentioned above, which will not be repeated here.

[0087] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program being implemented when executed by a processor:

[0088] Step 1: Acquire multi-channel helicopter acoustic signals using a microphone array and perform preprocessing;

[0089] Step 2: Perform phase-corrected frequency rearrangement synchronization compression transformation on the preprocessed helicopter acoustic signal;

[0090] Step 3: Perform an inverse short-time Fourier transform on the frequency domain result after the phase-corrected frequency rearrangement synchronous compression transform.

[0091] Step 4: Perform adaptive line spectrum enhancement on the multi-channel signals after inverse short-time Fourier transform to obtain the enhanced time-domain multi-channel helicopter acoustic signals.

[0092] For specific limitations on each step, please refer to the limitations of the helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transformation mentioned above, which will not be repeated here.

[0093] As a specific example, in one embodiment, the present invention relates to... Figure 2 The helicopter acoustic signal shown is enhanced by combining adaptive line spectrum enhancement and synchronous compression transform. Its time-frequency spectrum and power spectrum are as follows: Figure 4 As shown, the helicopter acoustic signal enhancement method of this invention, which combines adaptive line spectrum enhancement and synchronous compression transform, can improve the energy of each spectral line of the helicopter while suppressing noise and interference at frequencies surrounding the spectral lines, thereby significantly improving the narrowband signal-to-noise ratio of the helicopter acoustic signal. At different signal-to-noise ratios, DOA estimation was performed on the helicopter acoustic signal, the helicopter acoustic signal after phase-corrected frequency rearrangement synchronous compression transform, the helicopter acoustic signal after adaptive line spectrum enhancement, and the helicopter acoustic signal enhanced by combining adaptive line spectrum enhancement and synchronous compression transform, based on the MUSIC algorithm. The results are as follows. Figure 5 As shown in the figure, the present invention, by combining adaptive line spectrum enhancement and synchronous compression transformation algorithm, can improve the signal-to-noise ratio of helicopter acoustic signals under different signal-to-noise ratio conditions, thereby greatly improving the target direction finding accuracy of helicopters.

[0094] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention without departing from its spirit and scope should be included within the protection scope of the present invention.

Claims

1. A method for enhancing helicopter acoustic signals by combining adaptive line spectrum enhancement and synchronous compression transform, characterized in that, The method includes the following steps: Step 1: Acquire multi-channel helicopter acoustic signals using a microphone array and perform preprocessing. Step 2: Perform phase-corrected frequency rearrangement synchronization compression transformation on the preprocessed helicopter acoustic signal; Step 3: Perform an inverse short-time Fourier transform on the frequency domain result after the phase-corrected frequency rearrangement synchronous compression transform. Step 4: Perform adaptive line spectrum enhancement on the multi-channel signals after inverse short-time Fourier transform to obtain the enhanced time-domain multi-channel helicopter acoustic signals. Step 2, which involves performing phase-corrected frequency rearrangement synchronization compression transformation on the preprocessed helicopter acoustic signal, specifically includes: Step 2-1: Perform a windowed short-time Fourier transform on the helicopter acoustic signals of each channel to obtain: in, This represents the helicopter sound signal collected by the i-th microphone at sampling time m, i.e., the helicopter sound signal of the i-th channel at sampling time m. M represents the number of microphones in the microphone array, and g represents the Gaussian window function. denoted by , where * denotes the first derivative of the Gaussian window function, * denotes conjugate, and n represents the sampling time. N is the number of sampling points in one frame of the helicopter acoustic signal, and k represents the frequency. f s Sampling rate, This represents the result of taking the conjugate of the values ​​of the Gaussian window at sampling time mn. This represents the result of taking the conjugate of the first derivative of the Gaussian window at sampling time mn. Let represent the time-frequency representation of the helicopter acoustic signal of the i-th channel at the sampling time n and frequency k, after Fourier transform using a Gaussian window as the window function, at the instantaneous frequency point [n,k]. The time-frequency representation of the helicopter acoustic signal of the i-th channel at sampling time n and frequency k after Fourier transform using the first derivative of the Gaussian window as the window function; Step 2-2: For each time-frequency point of the helicopter acoustic signal in each channel, calculate its instantaneous frequency operator. This instantaneous frequency operator indicates the position of the energy centroid corresponding to each time-frequency point on the time-frequency plane. Among them, control threshold It is a positive number. This indicates taking the imaginary part; The frequency of the energy centroid at the sampling time n and frequency k after the Fourier transform of the helicopter acoustic signal of the i-th channel using a Gaussian window as the window function is represented. Steps 2-3, at sampling time n, represent the time-frequency of each time-frequency point as shown in equation (4), perform phase correction, and then rearrange it to its energy centroid position. Specifically, this includes: The time-frequency representation of the time-frequency point [n,k1] is as follows: The frequency of the energy centroid at that time point can be calculated according to equation (3). Then the time-frequency representation of its energy centroid location [n,k0] is as follows: ,in , and Let [n,k1] and [n,k0] be the amplitudes at the time-frequency points [n,k1] and [n,k0], respectively. and Let [n,k1] and [n,k0] be the phases at the time-frequency points [n,k1] and [n,k0], respectively. and These are the time delays at time-frequency points [n,k1] and [n,k0], respectively. right The following correction is made so that the phase of the time-frequency point [n,k1] becomes... : in, , , This indicates taking the phase; after time-frequency rearrangement, the time-frequency representation of the time-frequency point [n,k1] becomes... The time-frequency representation of its centroid location [n,k0] becomes .

2. The helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform according to claim 1, characterized in that, The preprocessing described in step 1 includes: Step 1-1: Eliminate the DC component from the helicopter's acoustic signal; Steps 1-2 involve frame segmentation.

3. The helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform according to claim 1, characterized in that, The control threshold Take 2.22 × 10 -16 .

4. The helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform according to claim 1, characterized in that, Step 3 describes performing an inverse short-time Fourier transform on the frequency domain result after the phase-corrected frequency rearrangement synchronous compression transform. The specific formula is as follows: in, The time-frequency representation at sampling time n and frequency k after phase correction and frequency rearrangement synchronous compression transformation. The value of the helicopter acoustic signal in the i-th channel at sampling time n after phase-corrected frequency rearrangement synchronous compression transformation.

5. The helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform according to claim 1, characterized in that, Step 4 involves performing adaptive line spectrum enhancement on the multi-channel signals after inverse short-time Fourier transform to obtain the enhanced time-domain multi-channel helicopter acoustic signal, as shown in the formula: in, To determine the value of the enhanced time-domain multi-channel helicopter acoustic signal at sampling time n, the input signal vector of the adaptive line spectrum enhancer... L is the filter length. , and These represent the values ​​of the time-domain helicopter acoustic signal of the i-th channel after phase-corrected frequency rearrangement synchronous compression transformation at sampling times n, n-1, and nL, respectively. The filter weight vector consists of the weights of each tap of the filter. composition, Filter weight vector .

6. The helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform according to claim 5, characterized in that, The filter weight vector is updated according to the following formula: Among them, step size factor This is used to control the increment magnitude of the filter tap weight vector; c is a constant used to avoid... Too small step size The value is too large, and the error signal at sampling time n , This is the expected value of the current input signal, i.e., the time-domain helicopter acoustic signal after phase correction, frequency rearrangement, synchronous compression transformation, at sampling time n.

7. The helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform according to claim 6, characterized in that, The step size factor The value range is (0,2).

8. The helicopter acoustic signal enhancement method combining adaptive line spectrum enhancement and synchronous compression transform according to claim 6, characterized in that, Delay the input signal The signal is fed into an adaptive filter. Expected value , where m is the number of sampling points with delay.

9. A helicopter acoustic signal enhancement system combining adaptive line spectrum enhancement and synchronous compression transform based on the method of any one of claims 1 to 8, characterized in that, The system includes: The first module is used to acquire multi-channel helicopter acoustic signals using a microphone array and to perform preprocessing. The second module is used to perform phase-corrected frequency rearrangement synchronous compression transformation on the preprocessed helicopter acoustic signal. The third module is used to perform inverse short-time Fourier transform on the frequency domain result after phase-corrected frequency rearrangement synchronous compression transform. The fourth module is used to perform adaptive line spectrum enhancement on the multi-channel signals after inverse short-time Fourier transform, so as to obtain the enhanced time-domain multi-channel helicopter acoustic signals.

Images