Leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio

Abstract

The invention provides a DEMON spectrum-based leaf frequency extraction method under a low signal-to-noise ratio. The method comprises the following steps: firstly, receiving an underwater target radiation noise signal by using a single hydrophone; then the modulation spectrum signal is demodulated through a square low-pass method, and a DEMON spectrum is obtained; secondly, background equalization is carried out on the DEMON spectrum through an alpha bidirectional filtering method, a line spectrum set K is preliminarily obtained, then line spectrum purification and leaf frequency extraction are carried out through different methods according to whether the axis frequency information is obtained or not, and finally the leaf frequency information can be accurately extracted under the low signal-to-noise ratio. Compared with a traditional leaf frequency extraction method, the leaf frequency can be accurately extracted at a lower signal-to-noise ratio, and a complex receiving array or an extraction algorithm is not needed; according to the method, different leaf frequency extraction methods can be selected according to whether the shaft frequency information is acquired or not.

Description

Technical Field

[0001] The invention belongs to the technical field of underwater acoustic target recognition, and in particular relates to a leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio, which is used for accurately extracting leaf frequency from passive sonar noise and improving target recognition capability. Background Art

[0002] With the increasing need for the development of the country's maritime military, the demand for underwater target feature extraction and identification is also increasing. The means of detecting maritime targets include infrared, radar, light, sound, etc. Since electromagnetic waves and light have a large attenuation in water and a limited propagation distance, acoustic signals are the main way to detect maritime targets. Among them, the noise signal radiated by underwater targets is the main source of information for target feature extraction and identification. When identifying target features, the most commonly extracted features are the shaft frequency, blade frequency, and number of blades of the target propeller. The relationship between the three is that the shaft frequency multiplied by the number of blades equals the blade frequency. Therefore, after obtaining the shaft frequency and blade frequency, the number of blades can also be obtained. The blade frequency contains both information about the shaft frequency and the number of blades, so it is crucial to accurately extract the blade frequency.

[0003] The traditional feature extraction method is to obtain the DEMON spectrum by demodulating the signal through the modulation spectrum, and extract the shaft frequency and blade frequency from the DEMON spectrum. For shaft frequency extraction, this method requires demodulating the DEMON spectrum from multiple high-frequency broadband signals for extraction, while for blade frequency extraction, demodulation in the low-frequency band is more effective. When extracting both at the same time, in order to accommodate shaft frequency extraction, the demodulated broadband signal needs to be improved, but at this time there are certain requirements for the signal-to-noise ratio of the signal, that is, the extraction cannot be completed under low signal-to-noise ratio. However, in practical applications, the radiated noise signal received by passive sonar often has a very low signal-to-noise ratio, so it is necessary to study a method to complete the feature extraction of underwater targets under low signal-to-noise ratio. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0005] To this end, the present invention provides a leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio, which can accurately extract leaf frequency under low signal-to-noise ratio.

[0006] The technical solution of the present invention is as follows:

[0007] According to one aspect, a leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio is provided, the extraction method comprising:

[0008] Step 1: Use a single hydrophone to receive the noise signal x(t) radiated by the underwater target;

[0009] Step 2: Obtain a DEMON spectrum according to the noise signal obtained in step 1;

[0010] Step 3: Perform background equalization on the DEMON spectrum through an α-bidirectional filter, and preliminarily obtain a line spectrum set K and an amplitude set Y corresponding to the line spectrum according to the filtering result and the set threshold Δ;

[0011] Step 4: determine whether the shaft frequency information is known, if so, go to step 5; if not, go to step 6;

[0012] Step 5: Substitute the shaft frequency information to purify the line spectrum set K, remove the interference line spectrum, retain the possible line spectrum of the leaf frequency, and obtain the line spectrum set K zp , and then substitute the axis frequency information, from the line spectrum K zp Extract leaf frequency from

[0013] Step 6: purify the line spectrum set K according to the line spectrum amplitude information, propose the interference line spectrum, retain the possible line spectrum of the leaf frequency, obtain the line spectrum set K0, and extract the leaf frequency from the line spectrum K0.

[0014] Furthermore, the step 2 specifically includes:

[0015] Select a low-frequency broadband filter to filter the noise signal, and then square the low-pass demodulated modulation spectrum signal to obtain the DEMON spectrum.

[0016] Furthermore, in step 2, the following signal is obtained after low-pass filtering:

[0017]

[0018] Where A is the amplitude of the signal; m is the modulation index, which satisfies: 0<m<1; ω is the carrier frequency; Ω is the modulation frequency; t is the time; the cutoff frequency of the low-pass filter is F lpf Satisfy: 2Ω<F lpf <2ω-2Ω; LF means low-pass filter;

[0019] The signal obtained after low-pass filtering is Fourier transformed to obtain the DEMON spectrum signal S DEMON .

[0020] Furthermore, the step three specifically includes:

[0021] The DEMON spectrum signal S DEMON As the input of α bidirectional filter, its adaptive output can obtain the continuous spectrum signal S of DEMON spectrum. 均衡 ;

[0022] According to S DEMON The signal amplitude sets the threshold Δ, and makes Δ×S 均衡 The amplitude is greater than S DEMON The continuous spectrum part, S DEMON -Δ×S 均衡Then, the negative values ​​are set to zero to obtain the valid line spectrum set K.

[0023] Furthermore, the α bidirectional filter is designed using the following formula:

[0024]

[0025]

[0026]

[0027] Where x(k) is the input of the α-bidirectional filter at time k. In step 3, the DEMON spectrum signal S DEMON Replace x(k) as the filter input, is the forward output of the filter at time k, is the inverse output of the filter at time k, and Q is the recursive coefficient.

[0028] Furthermore, the step five specifically includes:

[0029] 501 substitutes the axis frequency to perform line spectrum purification, including:

[0030] S1 sets the propeller blade number range according to the actual situation, multiplying it by the shaft frequency zp to obtain the blade frequency range [yp min ,yp max ], remove the line spectra that are not in this range in the line spectrum set K, and obtain a new line spectrum set K1, and the corresponding line spectrum amplitude set is Y1;

[0031] S2 calculates the fractional part of the line spectrum set K1 divided by the axial frequency zp: |K1 / zp-round(K1 / zp)|, and removes the part with a larger fractional part in the line spectrum to obtain a new line spectrum set K2, and the corresponding line spectrum amplitude set is Y2;

[0032] S3 merges the line spectra with adjacent frequency intervals less than 1 Hz in the line spectrum set K2, and adds the corresponding line spectrum amplitudes to obtain a new line spectrum set K3, and the corresponding line spectrum amplitude set is Y3;

[0033] In the S4 line spectrum set K3, the line spectra with adjacent frequency intervals less than 3Hz are removed according to the amplitude, and the line spectra with large amplitudes are retained, and finally the purified line spectrum K is obtained. zp , the corresponding line spectrum amplitude set is Y zp ;

[0034] 502 extracts leaf frequency according to the purified line spectrum, including:

[0035] S1 Calculation Line Spectrum K zp The frequency of the line spectrum with the largest amplitude is set as the temporary leaf frequency, that is, yp1 = K zp (max(Yzp ));

[0036] S2 calculates the line spectrum frequency K zp Divide by the fractional part of the shaft frequency zp: yp0 = |K zp / zp-round(K zp / zp)|, select the line spectrum corresponding to the minimum value in yp0 as the temporary leaf frequency, that is, yp2 = K zp (min(yp0));

[0037] S3 determines whether yp1 and yp2 are equal. If they are equal, yp=yp1=yp2; otherwise, yp=yp2. That is, the target leaf frequency is finally obtained as yp.

[0038] Furthermore, in step 501, the S4 further includes: if K zp With Y zp If it is an empty set, return to step 1 and receive the signal again.

[0039] Furthermore, the step six specifically includes:

[0040] 601 performs line spectrum purification according to the obtained line spectrum frequency set K, including:

[0041] S1 merges the line spectra with adjacent frequency intervals less than 1 Hz in the line spectrum set K, and adds the corresponding line spectrum amplitudes to obtain a new line spectrum set K01, and the corresponding line spectrum amplitude set is Y01;

[0042] In the S2 line spectrum set K01, the line spectra with adjacent frequency intervals less than 3Hz are removed according to the amplitude, and the line spectra with large amplitudes are retained to obtain a new line spectrum set K02, and the corresponding line spectrum amplitude set is Y02;

[0043] S3 removes the line spectrum with too small amplitude. The removal rule is determined by the ratio of the line spectrum amplitude to the total line spectrum amplitude. The line spectrum with a ratio less than 0.05 is removed, and finally the purified line spectrum K0 is obtained, and the corresponding line spectrum amplitude set is Y0;

[0044] 602 extracts leaf frequency according to the purified line spectrum, including:

[0045] S1 determines whether the number of line spectra is less than 3, if so, execute S2, otherwise execute S5;

[0046] S2 determines whether the number of line spectra is greater than 1. If not, the leaf frequency is equal to the unique line spectrum frequency, that is, yp = K0 (1). Otherwise, S3 is executed.

[0047] S3 determines whether the amplitude of the two line spectra Y0(1)>Y0(2) is established. If so, yp=K0(1), otherwise execute S4;

[0048] S4 determines whether |K0(2)-2K0(1)|<0.2 holds. If so, yp=K0(1), otherwise yp=K0(2);

[0049] S5 calculates the difference frequency cp of the line spectrum set K0 = K0(i+1)-K0(i), i = 1, 2, ..., N-1, N is the number of line spectra, and then counts the number of the same difference frequencies in the difference frequency cp and puts them in the set num. The corresponding set of difference frequency values ​​is cp0;

[0050] S6 determines whether all num are 1. If not, the leaf frequency is the difference frequency with the largest number, that is, yp = cp0 (max (num)). Otherwise, calculate cha i,j =K0(i)-cp0(j), i=1,2,…,N, j=1,2,…,M, where M is the number of different difference frequencies;

[0051] S7 judgment i,j <0.2, if yes, then yp=K0(i), otherwise calculate the maximum amplitude line spectrum frequency divided by the difference frequency cp0, that is, ys j =K0(max(Y0)) / cp0(j),j=1,2,…,M;

[0052] S8 judgment |ys j -round(ys j )|<0.1,j=1,2,…,M is established, if so, then yp=cp0(j), otherwise yp=K0(max(Y0)).

[0053] Furthermore, in step 601, S3 further includes: if K0 and Y0 are empty sets, returning to step 1 and receiving the signal again.

[0054] According to another aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0055] Compared with the traditional leaf frequency extraction method, the above technical solution does not extract synchronously with the shaft frequency, and can select a demodulation bandwidth that is more suitable for leaf frequency extraction, so that the leaf frequency can be accurately extracted at a lower signal-to-noise ratio without the need for a complex receiving array or extraction algorithm; the method of the present invention can select different leaf frequency extraction methods based on whether the shaft frequency information has been obtained. If the shaft frequency information is known, the leaf frequency extraction can be performed based on the shaft frequency information, and the extraction method is simpler. If the shaft frequency information is unknown, the leaf frequency extraction can be performed based on the information of the line spectrum itself, and a higher extraction accuracy rate can also be obtained. That is, compared with the prior art, the present invention has the following beneficial effects:

[0056] (1) Compared with the traditional method, the present invention can accurately extract leaf frequency at a very low signal-to-noise ratio;

[0057] (2) The present invention proposes two methods for purifying line spectra and extracting blade frequency for the two cases of known shaft frequency and unknown shaft frequency. This method of extracting blade frequency separately is not limited by shaft frequency extraction conditions when extracting blade frequency, and can compensate for shaft frequency information.

[0058] (3) The method proposed in the present invention is simple and easy to implement. It does not require array reception and processing, and can be done using a single hydrophone. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The included drawings are used to provide a further understanding of the embodiments of the present invention, which constitute a part of the specification, are used to illustrate the embodiments of the present invention, and together with the text description, explain the principles of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 It is the leaf frequency extraction flow chart of the present invention;

[0061] Figure 2 It is the flow chart of blade frequency extraction when shaft frequency is known;

[0062] Figure 3 This is the flow chart of blade frequency extraction when shaft frequency is unknown. DETAILED DESCRIPTION

[0063] It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0065] Unless otherwise specifically stated, the relative arrangement of the parts and steps described in these embodiments, numerical expressions and numerical values ​​do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be regarded as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0066] like Figure 1 As shown, in one embodiment of the present invention, a leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio is provided, and the extraction method comprises:

[0067] Step 1: Use a single hydrophone to receive the noise signal x(t) radiated by the underwater target;

[0068] Step 2: Obtain a DEMON spectrum according to the noise signal obtained in step 1;

[0069] Step 3: Perform background equalization on the DEMON spectrum through an α-bidirectional filter, and preliminarily obtain a line spectrum set K and an amplitude set Y corresponding to the line spectrum according to the filtering result and the set threshold Δ;

[0070] Step 4: determine whether the shaft frequency information is known, if so, go to step 5; if not, go to step 6;

[0071] Step 5: Substitute the shaft frequency information to purify the line spectrum set K, remove the interference line spectrum, retain the possible line spectrum of the leaf frequency, and obtain the line spectrum set K zp , and then substitute the axis frequency information, from the line spectrum K zp Extract leaf frequency from

[0072] Step 6: purify the line spectrum set K according to the line spectrum amplitude information, propose the interference line spectrum, retain the possible line spectrum of the leaf frequency, obtain the line spectrum set K0, and extract the leaf frequency from the line spectrum K0.

[0073] It can be seen that compared with the traditional leaf frequency extraction method, the embodiment of the present invention does not extract synchronously with the shaft frequency, and can select a demodulation bandwidth that is more suitable for leaf frequency extraction, so that the leaf frequency can be accurately extracted at a lower signal-to-noise ratio without the need for a complex receiving array or extraction algorithm; the method of the present invention can select different leaf frequency extraction methods according to whether the shaft frequency information has been obtained. If the shaft frequency information is known, the leaf frequency extraction can be performed based on the shaft frequency information, and the extraction method is simpler. If the shaft frequency information is unknown, the leaf frequency extraction can be performed based on the information of the line spectrum itself, and a higher extraction accuracy can also be obtained.

[0074] In some preferred embodiments, the specific steps of step 2 include:

[0075] A low-frequency broadband filter is set to perform bandpass filtering on the underwater target radiation noise signal x(t) received by a single hydrophone, and then perform square demodulation processing:

[0076]

[0077] In the formula, A is the amplitude of the signal; m is the modulation index, which satisfies: 0<m<1; ω is the carrier frequency; Ω is the modulation frequency. After low-pass filtering the above formula, we can get:

[0078]

[0079] Among them, the cutoff frequency of the low-pass filter is F lpf Satisfy the following relationship

[0080] 2Ω<F lpf <2ω-2Ω

[0081] After Fourier transform of the above signal, we can get the DEMON spectrum signal S DEMON .

[0082] In some preferred embodiments, step three specifically includes:

[0083] The obtained DEMON spectrum signal S DEMON The continuous spectrum signal S is obtained by using the α bidirectional filter 均衡 , where the design process of the α bidirectional filter is: x(k) is the input of the α bidirectional filter at time k, is the forward output of the filter at time k, is the reverse output of the filter at time k, Q is the recursive coefficient, generally in the form of an integer power of 2, and the expression of the α bidirectional filter is:

[0084]

[0085]

[0086]

[0087] The DEMON spectrum signal S DEMON Replace x(k) as the filter input, and its adaptive output can obtain the continuous spectrum signal S of the DEMON spectrum. 均衡 , according to S DEMON The signal amplitude sets the threshold Δ, so that Δ×S 均衡 The amplitude is greater than S DEMON The continuous spectrum part, S DEMON -Δ×S 均衡 Then, the negative values ​​are set to zero to obtain the valid line spectrum set K.

[0088] Reference Figure 2 As shown, in some preferred embodiments, step five specifically includes:

[0089] Substitute 501 into the axis frequency and perform line spectrum purification:

[0090] S1 is based on the actual setting of the propeller blade number range, multiplied by the shaft frequency zp, and the blade frequency range [yp min ,yp max ], remove the line spectra that are not in this range in the line spectrum set K, and obtain a new line spectrum set K1, and the corresponding line spectrum amplitude set is Y1;

[0091] S2 calculates the fractional part of the line spectrum set K1 divided by the shaft frequency zp, that is, |K1 / zp-round(K1 / zp)|. Since the blade frequency is in a multiple relationship with the shaft frequency, the shaft frequency divided by the blade frequency and its harmonic frequency should be close to an integer. Therefore, the part with a larger fractional part in the line spectrum is removed to obtain a new line spectrum set K2, and the corresponding line spectrum amplitude set is Y2.

[0092] S3 merges the line spectra with adjacent frequency intervals less than 1 Hz in the line spectrum set K2, and adds the corresponding line spectrum amplitudes to obtain a new line spectrum set K3, and the corresponding line spectrum amplitude set is Y3;

[0093] In the S4 line spectrum set K3, the line spectra with adjacent frequency intervals less than 3Hz are removed according to the amplitude, and the line spectra with large amplitudes are retained, and finally the purified line spectrum K is obtained. zp , the corresponding line spectrum amplitude set is Y zp , if K zp With Y zp If it is an empty set, return to step (1) and receive the signal again.

[0094] After obtaining the purified line spectrum at 502, the leaf frequency needs to be extracted based on it. The steps are as follows:

[0095] S1 Calculation Line Spectrum K zp The frequency of the line spectrum with the largest amplitude is set as the temporary leaf frequency, that is, yp1 = K zp (max(Yzp ));

[0096] S2 is the correction yp1, and the line spectrum frequency K is calculated zp The fractional part of the shaft frequency zp, that is, yp0 = |K zp / zp-round(K zp / zp)|, select the line spectrum corresponding to the minimum value in yp0 as the temporary leaf frequency, that is, yp2 = K zp (min(yp0));

[0097] S3 determines whether yp1 and yp2 are equal. If they are equal, yp=yp1=yp2; otherwise, yp=yp2. That is, the target leaf frequency is finally obtained as yp.

[0098] Reference Figure 3 As shown, in some preferred embodiments, step six specifically includes:

[0099] 601 performs line spectrum purification according to the obtained line spectrum frequency set K, and the process is as follows:

[0100] S1 merges the line spectra with adjacent frequency intervals less than 1 Hz in the line spectrum set K, and adds the corresponding line spectrum amplitudes to obtain a new line spectrum set K01, and the corresponding line spectrum amplitude set is Y01;

[0101] In the S2 line spectrum set K01, the line spectra with adjacent frequency intervals less than 3Hz are removed according to the amplitude, and the line spectra with large amplitudes are retained to obtain a new line spectrum set K02, and the corresponding line spectrum amplitude set is Y02;

[0102] S3 removes the line spectrum with too small amplitude. The removal rule is determined by the ratio of the line spectrum amplitude to the total line spectrum amplitude. The line spectrum with a ratio less than 0.05 is removed, and finally the purified line spectrum K0 is obtained. The corresponding line spectrum amplitude set is Y0. If K0 and Y0 are empty sets, return to step 1 and receive the signal again.

[0103] After obtaining the purified line spectrum at 602, the leaf frequency needs to be extracted based on it. The steps are as follows:

[0104] S1 determines whether the number of line spectra is less than 3, if so, execute S2, otherwise execute S5;

[0105] S2 determines whether the number of line spectra is greater than 1. If not, the leaf frequency is equal to the unique line spectrum frequency, that is, yp = K0 (1). Otherwise, S3 is executed.

[0106] S3 determines whether the amplitude of the two line spectra Y0(1)>Y0(2) is established. If so, yp=K0(1), otherwise execute S4;

[0107] S4 determines whether |K0(2)-2K0(1)|<0.2 holds. If so, yp=K0(1), otherwise yp=K0(2);

[0108] S5 calculates the difference frequency cp of the line spectrum set K0 = K0(i+1)-K0(i), i = 1, 2, ..., N-1, N is the number of line spectra, and then counts the number of identical difference frequencies in the difference frequency cp (for example, if there are 2 identical calculated difference frequencies, the number is recorded as 2, if there is no identical difference, the number is recorded as 1), and puts it in the set num, and the corresponding set of difference frequency values ​​is cp0;

[0109] S6 determines whether all the values ​​in num are 1. If not, the leaf frequency is the difference frequency with the largest number, that is, yp = cp0 (max (num)). Otherwise (that is, there is no line spectrum with the same difference frequency, then all are 1), calculate cha i,j =K0(i)-cp0(j), i=1,2,…,N, j=1,2,…,M, where M is the number of different difference frequencies;

[0110] S7 judgment i,j <0.2, if yes, then yp=K0(i), otherwise calculate the maximum amplitude line spectrum frequency divided by the difference frequency cp0, that is, ys j =K0(max(Y0)) / cp0(j),j=1,2,…,M;

[0111] S8 judgment |ys j -round(ys j )|<0.1,j=1,2,…,M is established, if so, then yp=cp0(j), otherwise yp=K0(max(Y0)).

[0112] In this step, those skilled in the art know that a line spectrum set refers to a line spectrum set whose line spectrum frequencies are arranged from small to large, that is, K0(i) refers to the ith line spectrum frequency, and the same is true for the corresponding amplitude.

[0113] According to another embodiment, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the method of the above embodiment is implemented when the processor executes the computer program.

[0114] A specific implementation example is given below.

[0115] The blade frequency extraction method based on DEMON spectrum under low signal-to-noise ratio designed in the embodiment of the present invention is verified by using simulation data. The radiation noise signals of four different surface and underwater targets are simulated and designed, and the target parameters are shown in Table 1.

[0116] Table 1 Target parameters

[0117] Target Type Number of propeller blades Propeller speed / r / s Cruising speed / kn Displacement / t surface boat 1 4 7 10 1313 surface boat 2 4 9 10 10000 Underwater Vehicle 1 5 6 5 82 Underwater Vehicle 2 7 1.82 20 5000

[0118] First, the parameters are given as follows: the broadband demodulation frequency is set to 10Hz~300Hz, the signal sampling rate is 5000Hz, the number of blades is in the range of 3~7, the signal-to-noise ratio is set to 10dB, 0dB, -10dB, and -15dB respectively, and the blade frequency extraction accuracy is calculated under the known shaft frequency and unknown shaft frequency conditions. The results are shown in Tables 2 and 3. It can be seen that when the shaft frequency is known, when the signal-to-noise ratio is not lower than -10dB, the blade frequency extraction accuracy of different targets is close to 100%, and when the signal-to-noise ratio drops to -15dB, the blade frequency extraction accuracy of different targets is about 90%; when the shaft frequency is unknown, when the signal-to-noise ratio is not lower than -10dB, the blade frequency extraction accuracy of different targets is greater than 95%, and when the signal-to-noise ratio drops to -15dB, the blade frequency extraction accuracy of different targets is about 70%.

[0119] Table 2 Simulation results of blade frequency extraction accuracy at different signal-to-noise ratios when shaft frequency is known

[0120] Signal-to-noise ratio (dB) 10 0 -10 -15 surface boat 1 100％ 100％ 99％ 88.5％ surface boat 2 100％ 99.5％ 99％ 91％ Underwater Vehicle 1 100％ 100％ 97.5％ 85.5％ Underwater Vehicle 2 100％ 99.5％ 99.5％ 90％

[0121] Table 3 Simulation results of blade frequency extraction accuracy under different signal-to-noise ratios when shaft frequency is unknown

[0122] Signal-to-noise ratio (dB) 10 0 -10 -15 surface boat 1 100％ 98.5％ 94％ 70.5％ surface boat 2 100％ 97.5％ 95％ 75.5％ Underwater Vehicle 1 99％ 97％ 95％ 70％ Underwater Vehicle 2 100％ 99％ 98.5％ 91.5％

[0123] In summary, the method proposed in the present invention can accurately extract the target leaf frequency at a lower signal-to-noise ratio. Compared with the prior art, the present invention has the following beneficial effects:

[0124] (1) Compared with the traditional method, the present invention can accurately extract leaf frequency at a very low signal-to-noise ratio;

[0125] (2) The present invention proposes two methods for purifying line spectra and extracting blade frequency for the two cases of known shaft frequency and unknown shaft frequency. This method of extracting blade frequency separately is not limited by shaft frequency extraction conditions when extracting blade frequency, and can compensate for shaft frequency information.

[0126] (3) The method proposed in the present invention is simple and easy to implement. It does not require array reception and processing, and can be done using a single hydrophone.

[0127] Features described and / or illustrated above for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or used in place of features in other embodiments.

[0128] It should be emphasized that the term "include / comprises" when used herein refers to the presence of features, integers, steps or components, but does not exclude the presence or addition of one or more other features, integers, steps, components or combinations thereof.

[0129] The above method of the present invention can be implemented by hardware, or by hardware combined with software. The present invention relates to such a computer-readable program, which, when executed by a logic component, enables the logic component to implement the above-mentioned device or component, or enables the logic component to implement the above-mentioned various methods or steps. The present invention also relates to a storage medium for storing the above program, such as a hard disk, a magnetic disk, an optical disk, a DVD, a flash memory, etc.

[0130] The many features and advantages of these embodiments are apparent from this detailed description, and thus the appended claims are intended to cover all such features and advantages of these embodiments that fall within their true spirit and scope. Furthermore, since numerous modifications and changes will readily occur to those skilled in the art, it is not intended that the embodiments of the invention be limited to the exact construction and operation illustrated and described, but rather all suitable modifications and equivalents falling within the scope thereof are intended to be covered.

[0131] Parts of the present invention that are not described in detail are well known to those skilled in the art.

Claims

1. A leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio, characterized in that: The extraction method comprises: Step 1: Use a single hydrophone to receive the noise signal x(t) radiated by the underwater target; Step 2: Obtain a DEMON spectrum according to the noise signal obtained in step 1; Step 3: Perform background equalization on the DEMON spectrum through an α-bidirectional filter, and preliminarily obtain a line spectrum set K and an amplitude set Y corresponding to the line spectrum according to the filtering result and the set threshold Δ; Step 4: determine whether the shaft frequency information is known, if so, go to step 5; if not, go to step 6; Step 5: Substitute the shaft frequency information to purify the line spectrum set K, remove the interference line spectrum, retain the possible line spectrum of the leaf frequency, and obtain the line spectrum set K zp , and then substitute the axis frequency information, from the line spectrum K zp Extract leaf frequency from Step 6: purify the line spectrum set K according to the line spectrum amplitude information, propose the interference line spectrum, retain the possible line spectrum of the leaf frequency, obtain the line spectrum set K0, and extract the leaf frequency from the line spectrum K0.

2. The leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to claim 1, characterized in that: The step 2 specifically includes: Select a low-frequency broadband filter to filter the noise signal, and then square the low-pass demodulated modulation spectrum signal to obtain the DEMON spectrum.

3. The leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to claim 2 is characterized in that: In the step 2, the following signal is obtained after low-pass filtering: Where A is the amplitude of the signal; m is the modulation index, which satisfies: 0<m<1; ω is the carrier frequency; Ω is the modulation frequency; t is the time; the cutoff frequency of the low-pass filter is F lpf Satisfy: 2Ω<F lpf <2ω-2Ω; LF means low-pass filter; The signal obtained after low-pass filtering is Fourier transformed to obtain the DEMON spectrum signal S DEMON .

4. The leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to claim 1, characterized in that: The step three specifically includes: The DEMON spectrum signal S DEMON As the input of α bidirectional filter, its adaptive output can obtain the continuous spectrum signal S of DEMON spectrum. 均衡 ; According to S DEMON The signal amplitude sets the threshold Δ, and makes Δ×S 均衡 The amplitude is greater than S DEMON The continuous spectrum part, S DEMON -Δ×S 均衡 Then, the negative values ​​are set to zero to obtain the valid line spectrum set K.

5. The leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to claim 4 is characterized in that: The α bidirectional filter is designed using the following formula: Where x(k) is the input of the α-bidirectional filter at time k. In step 3, the DEMON spectrum signal S DEMON Replace x(k) as the filter input, is the forward output of the filter at time k, is the reverse output of the filter at time k, and Q is the recursive coefficient.

6. A leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to any one of claims 1 to 5, characterized in that: The step five specifically includes: 501 substitutes the axis frequency to perform line spectrum purification, including: S1 sets the propeller blade number range according to the actual situation, multiplying it by the shaft frequency zp to obtain the blade frequency range [yp min ,yp max ], remove the line spectra that are not in this range in the line spectrum set K, and obtain a new line spectrum set K1, and the corresponding line spectrum amplitude set is Y1; S2 calculates the fractional part of the line spectrum set K1 divided by the axial frequency zp: |K1 / zp-round(K1 / zp)|, and removes the part with a larger fractional part in the line spectrum to obtain a new line spectrum set K2, and the corresponding line spectrum amplitude set is Y2; S3 merges the line spectra with adjacent frequency intervals less than 1 Hz in the line spectrum set K2, and adds the corresponding line spectrum amplitudes to obtain a new line spectrum set K3, and the corresponding line spectrum amplitude set is Y3; In the S4 line spectrum set K3, the line spectra with adjacent frequency intervals less than 3Hz are removed according to the amplitude, and the line spectra with large amplitudes are retained, and finally the purified line spectrum K is obtained. zp , the corresponding line spectrum amplitude set is Y zp ; 502 extracts leaf frequency according to the purified line spectrum, including: S1 Calculation Line Spectrum K zp The frequency of the line spectrum with the largest amplitude is set as the temporary leaf frequency, that is, yp1 = K zp (max(Y zp )); S2 calculates the line spectrum frequency K zp Divide by the fractional part of the shaft frequency zp: yp0 = |K zp / zp-round(K zp / zp)|, select the line spectrum corresponding to the minimum value in yp0 as the temporary leaf frequency, that is, yp2 = K zp (min(yp0)); S3 determines whether yp1 and yp2 are equal. If they are equal, yp=yp1=yp2; otherwise, yp=yp2. That is, the target leaf frequency is finally obtained as yp.

7. The leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to claim 6, characterized in that: In step 501, the step S4 further includes: if K zp With Y zp If it is an empty set, return to step 1 and receive the signal again.

8. A leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to any one of claims 1 to 5, characterized in that: The step six specifically includes: 601 performs line spectrum purification according to the obtained line spectrum frequency set K, including: S1 merges the line spectra with adjacent frequency intervals less than 1 Hz in the line spectrum set K, and adds the corresponding line spectrum amplitudes to obtain a new line spectrum set K01, and the corresponding line spectrum amplitude set is Y01; In the S2 line spectrum set K01, the line spectra with adjacent frequency intervals less than 3Hz are removed according to the amplitude, and the line spectra with large amplitudes are retained to obtain a new line spectrum set K02, and the corresponding line spectrum amplitude set is Y02; S3 removes the line spectrum with too small amplitude. The removal rule is determined by the ratio of the line spectrum amplitude to the total line spectrum amplitude. The line spectrum with a ratio less than 0.05 is removed, and finally the purified line spectrum K0 is obtained, and the corresponding line spectrum amplitude set is Y0; 602 extracts leaf frequency according to the purified line spectrum, including: S1 determines whether the number of line spectra is less than 3, if so, execute S2, otherwise execute S5; S2 determines whether the number of line spectra is greater than 1. If not, the leaf frequency is equal to the unique line spectrum frequency, that is, yp = K0 (1). Otherwise, S3 is executed. S3 determines whether the amplitude of the two line spectra Y0(1)>Y0(2) is established. If so, yp=K0(1), otherwise execute S4; S4 determines whether |K0(2)-2K0(1)|<0.2 holds. If so, yp=K0(1), otherwise yp=K0(2); S5 calculates the difference frequency cp of the line spectrum set K0 = K0(i+1)-K0(i), i = 1, 2, ..., N-1, N is the number of line spectra, and then counts the number of the same difference frequencies in the difference frequency cp and puts them in the set num. The corresponding set of difference frequency values ​​is cp0; S6 determines whether all num are 1. If not, the leaf frequency is the difference frequency with the largest number, that is, yp = cp0 (max (num)). Otherwise, calculate cha i,j =K0(i)-cp0(j), i=1,2,…,N, j=1,2,…,M, where M is the number of different difference frequencies; S7 judgment i,j <0.2, if yes, then yp=K0(i), otherwise calculate the maximum amplitude line spectrum frequency divided by the difference frequency cp0, that is, ys j =K0(max(Y0)) / cp0(j),j=1,2,…,M; S8 judgment |ys j -round(ys j )|<0.1,j=1,2,…,M is established, if so, then yp=cp0(j), otherwise yp=K0(max(Y0)).

9. The leaf frequency extraction method based on DEMON spectrum under low signal-to-noise ratio according to claim 8, characterized in that: In step 601, S3 further includes: if K0 and Y0 are empty sets, returning to step 1 and receiving the signal again.

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to claims 1 to 9 is implemented.

Images