CN117949894A - Target detection method, device, medium and equipment for ice environment - Google Patents
Target detection method, device, medium and equipment for ice environment Download PDFInfo
- Publication number
- CN117949894A CN117949894A CN202410188985.9A CN202410188985A CN117949894A CN 117949894 A CN117949894 A CN 117949894A CN 202410188985 A CN202410188985 A CN 202410188985A CN 117949894 A CN117949894 A CN 117949894A
- Authority
- CN
- China
- Prior art keywords
- sound pressure
- vibration velocity
- covariance matrix
- cross covariance
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 59
- 239000011159 matrix material Substances 0.000 claims abstract description 126
- 239000013598 vector Substances 0.000 claims abstract description 95
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 58
- 238000001228 spectrum Methods 0.000 claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 38
- 230000009466 transformation Effects 0.000 claims abstract description 37
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 18
- 230000003595 spectral effect Effects 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 36
- 238000012935 Averaging Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 2
- 238000004590 computer program Methods 0.000 claims description 2
- 239000000758 substrate Substances 0.000 claims 5
- 238000005516 engineering process Methods 0.000 description 14
- 238000010586 diagram Methods 0.000 description 8
- 230000002829 reductive effect Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 230000005855 radiation Effects 0.000 description 5
- 238000005070 sampling Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 5
- 230000001629 suppression Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 230000001427 coherent effect Effects 0.000 description 2
- 230000036961 partial effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 238000009394 selective breeding Methods 0.000 description 1
- 239000002023 wood Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a target detection method, device, medium and equipment for an ice environment, relates to the field of target detection, and aims to solve the problem of passive detection of a weak target in an ice complex sound field environment such as polar regions. The target detection method comprises the following steps: carrying out sound pressure and vibration velocity combined processing on the received signals of the vector circular array receiver to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point; performing modal domain transformation on the sound pressure vibration velocity cross covariance matrix to obtain a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver; according to the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver, toeplitz reconstruction and characteristic decomposition are carried out, and noise subspace signals corresponding to different scanning orientations are obtained; and obtaining a MUSIC spatial spectrum corresponding to the received signal by using a MUSIC algorithm according to the noise subspace signal, and determining a target azimuth based on a spectral peak of the MUSIC spatial spectrum. The target detection method provided by the invention realizes the passive detection of the weak target under ice.
Description
Technical Field
The present invention relates to the field of object detection technologies, and in particular, to a method, an apparatus, a medium, and a device for detecting an object in an ice environment.
Background
In an under-ice environment, especially in an under-ice environment in a polar sea area, the sound velocity distribution appears as a binaural sound velocity distribution characteristic. The dual-channel waveguide effect limits the propagation of ice source noise to the seabed, so that the noise at the offshore seabed is Gaussian, and the signal is received by being distributed on the seabed with obvious advantages. Firstly, only the condition that the noise is Gaussian noise is considered, the signal processing is easier, and secondly, the non-blind area detection in a certain distance can be realized by arranging the array near the seabed. However, the submarine array mode has limitations, when the target distance is far, the signal-to-noise ratio of submarine received signals is low due to large acoustic wave propagation loss, and when the traditional azimuth estimation method is used for processing, weak target signals are submerged in background noise and weak targets cannot be detected, so that the passive detection of the long-distance weak targets is a problem to be solved urgently.
The traditional Conventional Beam Forming (CBF) achieves the aim of canceling partial noise of signal homodromous addition by compensating the phase difference of signals received by different array elements, which is equivalent to spatial filtering and improves the signal-to-noise ratio of an output end. Firstly, calculating an autocorrelation matrix for a received array element domain signal x (t) to obtain a received signal cross covariance matrix R x, and then calculating a weighting vector w m (theta) of an mth array element under different orientations, wherein R represents a circular array radius, M represents an array element number, theta represents a scanning direction calculation, all orientations are required to be traversed, lambda represents wavelengths, and a conventional beam forming spatial spectrum is obtained by traversing all orientations theta calculationThe direction pointed by the maximum value of the spectrum peak is the direction of the target. The conventional beam forming method has strong robustness, but has the defects of wide main beam lobe, low spatial resolution and high beam sidelobe when the number of array elements is small, and has an unsatisfactory effect when the direction is measured aiming at a weak target.
The conventional beam forming method has the defects of wide main beam lobe, low spatial resolution, high side lobe and background, and easy inundation of a weak target, and is characterized in that: the spatial resolution of the conventional beam forming method depends on the array aperture, the array aperture is limited by the number of array elements and the wavelength corresponding to the highest frequency of the received signal, the space sampling can be deduced according to the Nyquist sampling theorem to meet the requirement that the distance between adjacent array elements is not more than half the wavelength corresponding to the highest frequency of the signal, otherwise, aliasing can occur on the spatial spectrum, in addition, the number of array elements cannot be infinitely many in engineering practice, so the array aperture is always limited, the conventional beam forming resolution is low, the limited number of array elements can cause side lobes and high background, and a weak target is easily submerged.
The MUSIC (multiple signal classification ) algorithm is to perform feature decomposition on the received data cross covariance matrix to obtain a noise subspace and to use orthogonality of the noise subspace and the signal subspace to realize direction finding. The received signal cross covariance matrix is subjected to characteristic decomposition to obtain K (K is the information source number) large characteristic values and M-K (M is the array element number) small characteristic values, the characteristic vectors corresponding to the K large characteristic values form a signal subspace, the characteristic vectors corresponding to the M-K small characteristic values form a noise subspace, the characteristic vectors corresponding to the K large characteristic values and the signal incident guide vector a (theta) are in the same subspace, so that the signal guide vector and the noise subspace E n are orthogonal, and the method is utilizedThe MUSIC spatial spectrum can be calculated, and the maximum value of the spectrum peak is the azimuth of the target. The MUSIC algorithm breaks through the limitation of array aperture, can realize high-resolution direction finding, but the performance of the algorithm is seriously degraded under the conditions of low signal-to-noise ratio and small snapshot, and is not suitable for direction finding of weak targets.
The MUSIC algorithm has the defects that the performance of the algorithm is reduced under the conditions of low signal-to-noise ratio and small-block shooting, and the direction finding error of a weak target is larger, which is caused by the following reasons: the MUSIC algorithm has the advantages that the performance of the MUSIC algorithm is reduced under the conditions of small snapshots and low signal to noise ratio, the calculated signal covariance and an ideal signal autocorrelation matrix have larger errors when the number of the snapshots is small, so that the algorithm performance is reduced, the signal subspace and the noise subspace are divided and fuzzy when the signal cross covariance matrix is decomposed under the condition of low signal to noise ratio, the obtained noise subspace and the obtained signal subspace do not meet strict orthogonality, the peak value is lower due to the fact that the signal guide vector and the noise subspace are not strong in orthogonality when the MUSIC spatial spectrum is calculated, the correlation between the signal guide vector and the noise subspace is not obvious, and the correlation between the signal guide vector and the background is not obvious, so that the MUSIC algorithm can not measure the weak target azimuth when carrying out azimuth estimation.
The method is characterized in that the method comprises the steps of arranging a passive sonar array at the offshore bottom far from an ice layer for realizing the passive detection of the target under the ice, and avoiding the adverse effect of the ice source noise on the passive detection. In summary, the present application aims to solve the problem of low signal to noise ratio detection in an ice environment, especially weak target detection.
Disclosure of Invention
The invention aims to solve the problem that the target passive detection in the ice complex sound field environment often cannot be detected, and provides a target detection technical scheme for the ice environment, in particular to a target passive detection method for the ice weak target. By fully playing the advantages of the array, combining the vector array and the circular array into a vector circular array and combining a weak target azimuth estimation method of a proper matched vector circular array, the passive detection of the weak target under ice is realized. In the under-ice environment, the vector circular array is utilized to detect the passive target, so that the problem that the target with low signal to noise ratio is difficult to detect is solved.
In order to achieve the above object, the present invention provides the following technical solutions:
a method of object detection for use in a sub-ice environment, comprising:
carrying out sound pressure and vibration velocity combined processing on the received signals of the vector circular array receiver to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point;
Performing modal domain transformation on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning directions to obtain a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver;
According to the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver, toeplitz reconstruction and characteristic decomposition are carried out, and noise subspace signals corresponding to different scanning orientations are obtained;
according to the noise subspace signals, MUSIC spatial spectrums corresponding to the received signals are obtained by using a MUSIC algorithm;
And determining the target azimuth according to the spectral peak of the MUSIC spatial spectrum.
Compared with the prior art, the target detection method for the ice environment provides a weak target azimuth estimation algorithm which organically combines the Toeplitz reconstruction technology, the circular array modal domain transformation technology, the vector array sound pressure vibration velocity combined processing method and the MUSIC method on the basis of the vector circular array aiming at the polar ice environment, and can realize accurate azimuth estimation of the ice long-distance weak target.
The Toeplitz matrix reconstruction method can reconstruct a cross covariance matrix of a received signal to Toeplitz, and can realize target azimuth estimation under low signal-to-noise ratio by combining with a MUSIC algorithm, but the Toeplitz reconstruction method is only applicable to linear arrays with array manifold meeting the Van der Mono matrix form, and for circular arrays, circular array modal domain transformation needs to be performed first, the circular array is transformed into a virtual linear array, and then Toeplitz treatment is performed on the cross covariance matrix.
The sound pressure and vibration velocity combined processing method of the vector array can better inhibit noise and improve signal to noise ratio, the x-channel and y-channel vibration velocity linear combination is used for constructing a combined vibration velocity Vc and combining the sound pressure channel to calculate a sound pressure and vibration velocity cross covariance matrix, isotropic noise is inhibited by utilizing the uncorrelation of noise between the vibration velocity channel and the sound pressure channel, and the direction estimation error is caused by artificial selection of the vibration velocity Vc observation direction.
Through sound pressure and vibration speed combined processing and Toeplitz reconstruction based on a modal domain, the direction estimation of a low signal-to-noise ratio weak target under ice can be realized.
The present invention also provides an object detection apparatus for an ice environment, comprising:
The joint processing unit is configured to perform sound pressure and vibration velocity joint processing on the received signals of the vector circular array receiver to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point;
The conversion unit is configured to perform modal domain conversion on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning directions to obtain a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver;
The reconstruction unit is configured to reconstruct Toeplitz and decompose features according to the sound pressure and vibration velocity cross covariance matrix of the virtual linear array receiver to obtain noise subspace signals corresponding to different scanning orientations;
The MUSIC processing unit is configured to obtain a MUSIC spatial spectrum corresponding to the received signal by utilizing a MUSIC algorithm according to the noise subspace signal;
And the azimuth identifying unit is configured to determine a target azimuth according to the spectral peak of the MUSIC spatial spectrum.
Compared with the prior art, the beneficial effects of the target detection device for the ice environment are the same as those of the target detection method for the ice environment in the technical scheme, and the description is omitted here.
The present invention also provides an object detection apparatus for use in an ice environment, comprising: a vector circular array receiver for receiving signals and an object detection device as described above.
The present invention also provides a computer storage medium having instructions stored therein which, when executed, implement the target detection method for an ice environment as described above.
The invention also provides a computing device comprising a processor and a communication interface coupled to the processor; the processor is configured to run a computer program or instructions to implement the object detection method for a sub-ice environment as described above.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
fig. 1 is a schematic diagram of a received signal model of a vector circular array receiver according to an embodiment of the present invention.
Fig. 2 is a flow chart of a method for detecting an object in an ice environment according to an embodiment of the invention.
Fig. 3 is a schematic block diagram of an object detection device for use in an ice environment according to an embodiment of the present invention.
Fig. 4 is a schematic flow chart of receiving signals from a vector circular array until a MUSIC spatial spectrum is output in an embodiment of the invention.
Fig. 5 is a schematic diagram showing the comparison of the output spatial spectrum of the target detection scheme and the existing multiple target detection algorithms in the embodiment of the present invention.
Fig. 6 is a comparison diagram of root mean square error of azimuth estimation of each algorithm in the embodiment of the invention.
FIG. 7 is a comparison diagram of average background levels of the orientation estimation of each algorithm in an embodiment of the present invention.
Fig. 8 is a schematic diagram of a spatial spectrum comparison result obtained by fixing the observation direction of the combined vibration velocity Vc and matching the observation direction of the combined vibration velocity Vc with the scanning direction in the embodiment of the present invention.
Detailed Description
In order to clearly describe the technical solution of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", etc. are used to distinguish the same item or similar items having substantially the same function and effect. For example, the first threshold and the second threshold are merely for distinguishing between different thresholds, and are not limited in order. It will be appreciated by those of skill in the art that the words "first," "second," and the like do not limit the amount and order of execution, and that the words "first," "second," and the like do not necessarily differ.
In the present invention, the words "exemplary" or "such as" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.
In the present invention, "at least one" means one or more, and "a plurality" means two or more. "and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a alone, a and B together, and B alone, wherein a, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b or c may represent: a, b, c, a and b, a and c, b and c, or a, b and c, wherein a, b, c can be single or multiple.
In the technical field of under-ice target detection, compared with a sound pressure hydrophone, the vector hydrophone has sound pressure and a plurality of vibration velocity channels, can acquire more target information, and can obtain higher space gain and stronger noise suppression capability by utilizing the difference of sound pressure and particle vibration velocity in a target radiation noise field and an environment noise field in space correlation. Compared with a linear array, the circular array has approximate azimuth estimation precision and spatial resolution in each azimuth, and engineering layout is simpler in practice. The spatial resolution of the uniform circular array is very similar in the scanning range of 0-360 degrees, but the uniform linear array is only higher in the spatial resolution of 60-120 degrees in the scanning range of 0-180 degrees, and the closer to 90 degrees, the higher the spatial resolution is, and the closer to 0-180 degrees, the lower the spatial resolution is.
The vector circular array receiver comprises M array elements, wherein the M array elements are vector array elements, and the vector circular array receiver is formed by the M array elements together. The vector circular array receiver is a vector circular array passive sonar array. The grating lobes appear on the spatial spectrum, which can bring adverse effect to the azimuth estimation, in order to avoid the grating lobes, the spacing between two adjacent array elements is required to be less than half of the wavelength corresponding to the highest frequency of the received signal when the passive sonar array is deployed. Based on the above, the relation between the array element number M of the circular array and the radius R of the circular array and the wavelength lambda corresponding to the highest frequency of the received signal can be deduced, and the requirement is satisfied
For a uniform vector circular array, the vector is compared with a scalar, namely a sound pressure array; the circular array is an array type of an array, and the array type of the array is numerous and comprises a linear array, an arc array, a cross array and the like. By uniform is meant that the array elements are equally spaced, as opposed to non-uniform, sparse, etc.
As shown in fig. 1, a received signal model of a vector circular receiver is given. The vector circular array with the array element number M is positioned in an xOy plane, the radius is R, and the positive direction of the x axis is taken as the 0-degree direction. A vector circular array is an array of many array elements, a single array element being called a vector hydrophone or acoustic vector sensor. The vector hydrophones include acoustic pressure hydrophones and/or velocity hydrophones, including, for example, one acoustic pressure hydrophone, two or three velocity hydrophones. The elements for acquiring sound pressure are sound pressure hydrophones, and the elements for acquiring x and y vibration speeds are an x vibration speed hydrophone and a y vibration speed hydrophone respectively. The vector sensor includes a sound pressure sensor (i.e., a sound pressure hydrophone) and a vibration velocity sensor. The vibration velocity sensor includes a displacement sensor, a velocity sensor, an acceleration sensor, etc., and may be referred to as a vector hydrophone although the principle is similar.
The channel directions of vibration velocity components x and y of the vector circular array are respectively consistent with the directions of an x axis and a y axis. S represents an incident information source, and the information source is mainly a target ship radiation noise signal in a passive sonar detection mode, which is a broadband signal, and main frequency components are concentrated below 1 kHz; x, y, z, O three symbols form a space rectangular coordinate system, which is used herein to describe a three-dimensional space; θ represents an included angle between a straight line where the projection of the incident direction of the information source S on the xOy plane and the positive direction of the x axis, and the included angle is determined by an angle from the positive direction of the x axis to the direction of the projection in a way of anticlockwise rotation, and ranges from 0 degree to 360 degrees; the included angle between the incident direction of the information source S and the positive direction of the z axis is in the range of 0-180 degrees.
The scanning direction refers to the scanning azimuth or scanning direction angle of the vector circular array, and the observation direction refers to the direction of the combined vibration velocity Vc. The observation direction is essentially the electronic rotation direction of the vector hydrophone, the x-channel vibration velocity and the y-channel vibration velocity of the vector hydrophone can be linearly combined through sine and cosine values of the electronic rotation angle to obtain a combined vibration velocity Vc, so that the combined vibration velocity Vc has a direction, and the direction angle is essentially the electronic rotation angle of the vector hydrophone and is uniformly defined as an observation direction and is denoted by a symbol phi. In the present application, the observation direction is creatively consistent with the scanning direction, and the observation direction is changed along with the change of the scanning direction.
As shown in fig. 2, a flow chart of a target detection method for an ice environment is provided, which includes the following steps:
s1, carrying out sound pressure and vibration velocity combined processing on a received signal of a vector circular array receiver to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point;
S2, performing modal domain transformation on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning directions to obtain a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver;
s3, performing Toeplitz reconstruction and feature decomposition according to a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver to obtain noise subspace signals corresponding to different scanning orientations;
step S4, obtaining a MUSIC spatial spectrum corresponding to the received signal by utilizing a MUSIC algorithm according to the noise subspace signal;
and S5, determining a target azimuth according to the spectrum peak of the MUSIC spatial spectrum.
The sound pressure and vibration velocity combined processing for the received signals of the vector circular array receiver comprises the following steps:
Constructing a broadband receiving signal model based on a vector circular array under ice to obtain a sound pressure receiving signal and x, y vibration speed receiving signals;
Combining the vibration speed signal in the x direction and the vibration speed signal in the y direction to obtain a combined vibration speed signal;
performing cross-correlation processing on the combined vibration velocity signal and the sound pressure signal at each frequency point to obtain a sound pressure vibration velocity cross-covariance matrix at each frequency point;
And traversing the combined vibration velocity in the range of 0-360 degrees in an omnidirectional manner to obtain the sound pressure in each scanning direction and the cross covariance matrix of the combined vibration velocity under different frequency points.
The combined vibration speed observation direction phi is consistent with the scanning direction of the vector circular array.
The method for performing modal domain transformation on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning orientations into the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver comprises the following steps:
Carrying out Fourier transform processing on the received signals of the vector circular array;
Determining a modal transformation matrix according to the signal after the space domain discrete Fourier transformation;
Based on the modal transformation matrix, carrying out modal domain transformation on the sound pressure vibration velocity cross covariance matrix under each scanning direction of different frequency points, and respectively outputting the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver obtained by the modal domain transformation.
The Toeplitz reconstruction and characteristic decomposition are carried out according to the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver, and the method comprises the following steps:
time-averaging the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver to obtain the sound pressure vibration velocity cross covariance matrix of each scanning direction under different frequencies;
Sequentially carrying out frequency averaging on the cross covariance matrixes of different frequency points under each scanning azimuth to obtain a frequency-averaged cross covariance matrix;
performing Toeplitz reconstruction on the cross covariance matrix after frequency averaging;
And performing feature decomposition on the cross covariance matrix obtained by Toeplitz reconstruction to obtain noise subspaces corresponding to different scanning orientations.
As shown in fig. 3, a schematic diagram of an object detection device 100 for use in a sub-ice environment is provided, comprising:
A joint processing unit 101, configured to perform sound pressure and vibration velocity joint processing on the received signal of the vector circular array receiver, so as to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point;
The transformation unit 102 is configured to perform modal domain transformation on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning orientations to obtain a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver;
A reconstruction unit 103, configured to perform Toeplitz reconstruction and feature decomposition according to the acoustic pressure and vibration velocity cross covariance matrix of the virtual linear array receiver, so as to obtain noise subspace signals corresponding to different scanning orientations;
A MUSIC processing unit 104 configured to obtain a MUSIC spatial spectrum corresponding to the received signal by using a MUSIC algorithm according to the noise subspace signal;
An orientation recognition unit 105 configured to determine a target orientation from a spectral peak of the MUSIC spatial spectrum.
The following describes the technical scheme of wood strip detection according to the present application in detail with reference to the accompanying drawings.
In step S1, the received signal of the vector circular array receiver is subjected to sound pressure and vibration velocity joint processing, so as to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point. The step S1 specifically comprises the following steps:
S11, constructing a broadband receiving signal model based on a vector circular array under ice to obtain a sound pressure receiving signal P (f j) and x and y vibration speed receiving signals V x(fj)、Vy(fj);
In the passive sonar detection mode, the source is a target ship radiation noise broadband signal, the broadband signal is used for carrying out azimuth estimation, the signal is segmented in the time domain and used for increasing the frequency domain snapshot number (the time domain segmentation number corresponds to the frequency domain snapshot number), and the MUSIC method performance is poor due to the fact that the snapshot number is too small.
Assuming that H uncorrelated far-field wideband signals are incident on a vector circular array, dividing the received wideband signals into L segments, and performing J-point (J is the number of time-domain snapshots of each segment of the signal) discrete fourier transform on each segment of the wideband signals, where the received signal at frequency f j is represented as:
In equation 1, j=1, 2,..j, P (f j) represents the sound pressure signal received at frequency f j, V x(fj) and V y(fj) represent the received vibration velocity signals of the x and y channels at frequency f j, respectively, which together constitute the received signal of the vector circular array. A (f j, θ) represents a steering vector of the sound pressure channel at the frequency f j, N p(fj represents a sound pressure signal of noise, S (f j) represents a target source radiation noise signal at the frequency f j, N x(fj) and N y(fj) represent noise vibration velocity signals of the x and y channels, respectively, and θ represents a direction of an incident signal.
S12, combining the vibration speed signal V x(fj in the x direction and the vibration speed signal V y(fj in the y direction to obtain a combined vibration speed signal V c(fj)＝Vx(fj)cosφ+Vy(fj) sin phi, wherein phi is the combined vibration speed observation direction. The combined vibration speed observation direction phi is consistent with the scanning direction of the vector circular array, and detection errors caused by the preset phi direction are avoided. If the combined vibration velocity observation direction phi adopts a preset value, the preset value deviates from the true target azimuth, and a direction finding error is caused.
S13, performing cross-correlation processing on the combined vibration velocity signal V c(fj and the sound pressure signal P (f j) at each frequency point to obtain a sound pressure vibration velocity cross-covariance matrix at each frequency point
The sound pressure signal and the vibration velocity signal of far-field target radiation noise are coherent, and the signal coherence means that the two signals meet the following conditions in an meeting area: ① The vibration directions are the same; ② The vibration frequency is the same; ③ The phase is the same or the phase difference remains constant.
For an isotropic noise field, the sound pressure signal of the noise of the co-point synchronous measurement is uncorrelated with the vibration speed signal, and the co-point synchronous measurement means that two signals meet at a measurement point, and the time when the two signals reach the measurement point is consistent and the positions are the same. (the measurement point may be arbitrary). The sound pressure and vibration velocity of the noise signals measured on the different array elements are not relevant.
The vector hydrophones are single array elements forming a vector circular array, and 1 set of vector hydrophones comprises 1 sound pressure hydrophone and 2-3 vibration velocity hydrophones. And the irrelevance of the sound pressure and vibration speed of the noise signal in the isotropic noise field by the vector hydrophone is utilized to carry out sound pressure and vibration speed combined information processing, thereby inhibiting isotropic noise and fully playing the detection performance of the vector array.
S14, traversing the combined vibration velocity observation direction phi in the range of 0-360 degrees in an all-direction mode to obtain the sound pressure in each scanning direction and the cross covariance matrix of the combined vibration velocity under different frequency points. Sound pressure vibration velocity cross covariance matrix under each frequency pointExpressed as:
Wherein E (·) represents the average value of the matrix, (P+Vc) represents the addition of the data matrix P of the sound pressure channels of all the array elements and the calculated combined vibration velocity data matrix Vc of all the array elements, and (-) H represents the conjugate transpose of the calculated matrix. Matrix array M is the number of array elements; p represents a matrix formed by sound pressure channel data of all array elements, assuming M rows and N columns (N represents snapshot number), vc represents a matrix formed by combining vibration speed data of all array elements, and the matrix is also formed by multiplying each column in M rows and N columns, (P+Vc) and each row in (P+Vc) H, so that an M×M matrix can be obtained, N M×M matrices are obtained in total, and the average of the N matrices is calculated to be/>
Processing each segment in the received signal divided into L segments according to steps S12-S13 to obtain a sound pressure vibration velocity cross covariance matrix under each frequency point
In step S2, performing mode domain transformation on the sound pressure velocity cross covariance matrix of each frequency point under different scanning orientations, so as to obtain the sound pressure velocity cross covariance matrix of the virtual linear array receiver.
Calculating the sound pressure vibration velocity cross covariance matrix firstly needs to obtain a combined vibration velocity Vc, wherein the combined vibration velocity Vc has an observation direction, the direction estimation error is caused by manually selecting the observation direction, and the error can be avoided by keeping the observation direction consistent with the scanning direction of the array, but the process is at the cost of increasing the calculation amount. It should be noted, therefore, that here a sound pressure velocity cross-covariance matrix "at different scan orientations" is calculated. For example, the scanning direction is changed from 0 ° to 360 ° with 1 ° intervals, and then the sound pressure velocity cross covariance matrix corresponding to each 1 ° in the range of 0 ° to 360 ° needs to be calculated here.
When H far-field incidence sources are incident to the vector circular array in different directions, the receiving signal of the M-th array element of the vector circular array with the M-th array element is expressed as follows:
In formula 2, j represents an imaginary unit, x m represents a received signal of an mth array element, θ h represents an incident angle of an h signal, and s h represents a source signal of the h incident signal.
The step S2 specifically comprises the following steps:
S21, performing space domain discrete Fourier transform processing on the received signal of the m-th array element of the vector circular array to obtain a signal v i after the Fourier transform processing, wherein the signal v i is expressed as follows:
In the above formula, J represents an imaginary number, h=1, 2,..h, v i are the spatial discrete fourier transformed signals of x m, J -i (·) represents a bessel function of the-i-th order, and K is a constant.
S22, determining a modal transformation matrix according to the signals after the space domain discrete Fourier transformation.
When the number M of array elements meets the condition: time,/> ([. Cndot. ] represents rounding).
Equation 3 is now reduced to:
let u i＝vi rewrite the above formula into the following matrix form:
formula 5 is abbreviated as Is the array manifold of the virtual linear array obtained after the mode domain transformation.
The spatial discrete fourier transform according to equation 3 can be seen:
Order the
Equation 6 can be reduced to:
equation 8 can be abbreviated as u=f H x, combined with the foregoing The method can obtain the following steps:
In equation 9, x is the vector circular array received signal of M array elements. Order the The modality conversion matrix T is expressed as:
Wherein M is the number of array elements of the vector circular array, and J and F both represent the matrix, and are calculated as follows:
in the above formula, J represents an imaginary number, J -K (·) represents a-K-th order bessel function, and R represents a vector circular array radius.
The output signal of the vector circular array may be transformed into a "virtual linear array" output signal:
Wherein the method comprises the steps of
Equation 12 represents the array manifold of the "virtual linear array" obtained after the modal domain transformation.
S23, performing modal domain transformation on the sound pressure vibration velocity cross covariance matrix under each scanning direction of different frequency points based on the modal transformation matrix to respectively obtain a cross covariance matrix after the modal domain transformation as
Where f j denotes the corresponding frequency and θ denotes the scanning orientation.
The cross covariance matrix after the modal domain transformation is:
The output signal is a virtual linear array, and the sound pressure vibration velocity cross covariance matrix after the mode conversion is changed into R e(fj and theta).
In step S3, toeplitz reconstruction and feature decomposition are performed according to the acoustic pressure velocity cross covariance matrix of the virtual linear array receiver, so as to obtain noise subspace signals corresponding to different scanning orientations. The step S3 specifically comprises the following steps:
S31, performing time average on a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver to obtain a processed cross covariance matrix R eL(fj and theta); the method comprises the following steps: performing time average on the mode domain transformation of the sound pressure vibration velocity cross covariance matrix under each scanning direction of different frequency points to obtain a processed cross covariance matrix R eL(fj and theta);
S32, respectively calculating cross covariance matrixes after L-section signal modal domain transformation, and carrying out time averaging to obtain cross covariance matrixes R eL(fj and theta of each scanning direction under different frequencies;
S33, sequentially carrying out frequency averaging on cross covariance matrixes R eL(fj and theta of different frequency points under each scanning azimuth;
The cross covariance matrix R eL(fj, θ) of each scanning azimuth different frequency after the mode transformation and time averaging is subjected to the following frequency averaging
In the above-mentioned method, the step of,The number of frequency points corresponding to the upper limit frequency f h of the processing band is represented, and [ · ] represents a rounding. f s denotes the sampling rate, J denotes the discrete Fourier points,/>The number of frequency points corresponding to the lower limit frequency f l of the processing band is represented, and R eLf (θ) represents the cross covariance matrix at different scanning orientations after different frequency de-averaging.
S34, performing Toeplitz reconstruction on the cross covariance matrix after frequency averaging;
Under ideal conditions, the uniform linear array receiving data cross covariance matrix is a Toeplitz matrix, the ideal cross covariance matrix of the virtual linear array obtained by the circular array after the modal domain transformation also accords with the Toeplitz matrix form, and the matrix R eLf (theta) is obviously not the Toeplitz matrix form under the condition of low signal to noise ratio, and Toeplitz reconstruction is needed. The ideal Toeplitz matrix R 0 satisfies the following form
The main diagonal of the cross covariance matrix R eLf (θ) is subjected to Toeplitz processing on elements parallel to the main diagonal. For processed matricesThe process of Toeplitz reconstruction is represented by the following formula
Wherein M is the number of array elements, k=m-n. r mn Respectively represent R and/>M-th, n-th column element of (c).
S35, reconstructing Toeplitz to obtain a cross covariance matrixAnd performing feature decomposition to obtain noise subspaces E n (theta) corresponding to different scanning orientations.
Cross covariance matrix after Toeplitz reconstructionAnd performing feature decomposition, and selecting feature vectors corresponding to the noise feature values to obtain noise subspaces E n (theta) corresponding to different scanning orientations.
And sorting the characteristic values obtained after the characteristic decomposition according to the size, selecting a corresponding number of information source characteristic values for the information sources according to the sequence from large to small, and taking the rest characteristic values as noise characteristic values. The total number of the characteristic values is equal to the number of the array elements, and the number of the characteristic values of the information sources is consistent with the number of the information sources. For example, the number of array elements is 8, the number of sources (target number) is 2, then the characteristic values are sorted from big to small, the first 2 characteristic values are sequentially selected as the source characteristic values, and the remaining 6 characteristic values are used as the noise characteristic values. And obtaining the noise subspaces corresponding to different scanning orientations according to the feature vectors corresponding to the selected noise feature values.
In step S4, the noise subspace E n (θ) obtained in S3 under different scanning orientations is brought into the calculation expression of the output power of the MUSIC algorithm:
Where a p (θ) represents the sound pressure array scan vector and E n (θ) represents the noise subspace at different scan orientations.
Normalizing the MUSIC output power of all scanning orientations and taking the logarithm to obtain a spatial spectrum P MUSIC-out (theta) as follows:
In the above formula, P MUSIC-out (θ) represents the output spatial spectrum, and max (·) represents the maximum value.
In step S5, after the spatial spectrum output by the MUSIC is obtained, the target azimuth is determined according to the spectral peak. For example, two spectral peaks appearing in the directions of-30 ° and 60 ° in fig. 5 are the orientations of two targets.
As shown in fig. 4, a schematic flow chart of the spatial spectrum of the MUSIC from the vector circular array reception signal to the output MUSIC is given. Sequentially carrying out time domain segmentation on the received signals from the vector circular array; performing a Fast Fourier Transform (FFT) on the received signal in each time domain segment; calculating a sound pressure vibration velocity cross covariance matrix according to the frequency division point sub scanning azimuth; performing modal transformation on the sound pressure vibration velocity cross covariance matrix; cross covariance matrix time average of each time period of the frequency division point sub-scanning azimuth and cross covariance matrix frequency average of different frequency points of the sub-scanning azimuth; performing Toeplitz reconstruction and feature decomposition on the frequency-averaged cross covariance matrix to obtain a noise subspace; obtaining a MUSIC spatial spectrum by using a MUSIC algorithm according to the noise subspace; a target bearing (the target bearing representing a horizontal angle of incidence) is determined from the spatial spectrum.
As shown in fig. 5, a schematic diagram of the output spatial spectrum comparison of the target detection scheme of the present application with the existing multiple target detection algorithms is given. In the simulation process, a vector circular array with radius of 0.75M and 16 (M=16) array elements is arranged, independent double-target broadband signals are incident from-30 degrees and 60 degrees, the in-band signal-to-noise ratio is-10 dB and-15 dB respectively, and the signal duration is 1s. The time domain signal is divided into 100 (L=100) segments for processing, the sampling frequency is 32kHz, the FFT point number is 320, the scanning azimuth range is 0-360 degrees, and the angle interval is 1 degree.
The target detection method is marked as Toeplitz reconstruction mode domain (P+V c)*(P+Vc)H MUSIC), and the algorithm for comparison comprises array element domain sound pressure MUSIC, array element domain vector independent channel MUSIC, array element domain (P+V c)*(P+Vc)H MUSIC and mode domain (P+V c)*(P+Vc)H MUSIC).
According to the Toeplitz reconstruction mode domain (P+V c)*(P+Vc)H MUSIC) provided by the application has the advantages that the background level is the lowest, the weak target azimuth can be accurately measured, and the spectrum peak is the narrowest.
As shown in FIG. 6, the Root-mean-square Error (RMSE) of each algorithm orientation estimate is given, the Root-mean-square Error being defined as the orientation estimateSquare root of the ratio of sum of squares of deviations from true azimuth value θ to number of estimations χ, i.e.:
In the simulation of fig. 6, a vector circular array with 16 array elements and a radius of 0.75m is set, a single target broadband signal is incident from a 60-degree direction, and the acoustic pressure MUSIC of an array element domain, the vector independent channel MUSIC of the array element domain, the array element domain (P+V c)*(P+Vc)H MUSIC, the modal domain (P+V c)*(P+Vc)H MUSIC) and the root mean square error of the Toeplitz reconstruction modal domain (P+V c)*(P+Vc)H MUSIC algorithm) of the application under different signal to noise ratios are respectively compared, so that when the signal to noise ratio is higher, the estimated root mean square error of the Toeplitz reconstruction modal domain (P+V c)*(P+Vc)H MUSIC algorithm and other algorithms is very close, and when the signal to noise ratio is lower, the estimated root mean square error of the Toeplitz reconstruction modal domain (P+V c)*(P+Vc)H MUSIC algorithm) of the application is higher than the vector independent channel MUSIC of the array element domain, and the modal domain (P+V c)*(P+Vc)H MUSIC, the modal domain (P+V c)*(P+Vc)H MUSIC algorithm) of the application is slightly within a 3-degree range.
Fig. 6 shows that the Toeplitz reconstruction mode domain (p+v c)*(P+Vc)H MUSIC algorithm) of the present application has the advantages of high resolution and low background noise compared with other algorithms, and at the same time, still has higher estimation accuracy.
As shown in fig. 7, the average background level of each algorithm azimuth estimate is shown, the average background level (Average Background Level, ABL) being defined as the average of the normalized spatial spectrum (expressed logarithmically) except for the corresponding angular range of the first side lobes on both sides of the target azimuth, expressed by:
In the above formula, θ min is the minimum value of the scanning angle, and θ max is the maximum value of the scanning angle; θ v1 is the corresponding angle of the first side lobe at the left side of the target azimuth, and θ v2 is the corresponding angle of the first side lobe at the right side of the target azimuth; num θ is the number of scan angle points excluding the range enclosed by the first side lobe. The average background level can represent the characteristics of the side lobes of the spatial spectrum, reflecting the suppression capability of the algorithm to noise interference and the capability of improving the signal to noise ratio.
In the simulation of fig. 7, a vector circular array with a radius of 0.75m of 16 elements is set, a single-target broadband signal is incident from a 60-degree direction, and the acoustic pressure MUSIC of an array element domain, an array element domain vector independent channel MUSIC, an array element domain (p+v c)*(P+Vc)H MUSIC, a modal domain (p+v c)*(P+Vc)H MUSIC) and an average background level of the Toeplitz reconstruction modal domain (p+v c)*(P+Vc)H MUSIC algorithm) of the application under different signal-to-noise ratios are respectively compared, so that the average background level of each algorithm is raised along with the reduction of the signal-to-noise ratio, and the average background level of the Toeplitz reconstruction modal domain (p+v c)*(P+Vc)H MUSIC algorithm) provided by the application is far lower than that of other algorithms as the signal-to-noise ratio is raised, which indicates that the algorithm provided by the application can reduce the spatial spectral background to be very low, reduce the possibility that a weak target with a low signal-to-noise ratio is submerged in background noise, and has better weak target detection capability.
Fig. 8 shows the spatial spectrum comparison result of the algorithm of the present application for two cases of the fixed observation direction of the combined vibration velocity V c and the matching of the observation direction of the combined vibration velocity V c to the scanning direction. A vector circular array with a radius of 0.75m of 16 elements is arranged in simulation, independent double-target broadband signals are incident from-30 degrees and 60 degrees, the in-band signal-to-noise ratio is-10 dB and-15 dB respectively, the signal duration is 1s, the time domain is processed in 100 sections, the sampling rate is 32kHz, the FFT point number is 320, the scanning azimuth range is 0-360 degrees, the angle interval is 1 degree, the spatial spectrums of the algorithm under the two conditions that the observation direction of the combined vibration velocity V c is fixed and the observation direction of the combined vibration velocity V c is matched with the scanning direction are respectively compared, through comparison, the observation result of the combined vibration velocity V c is 66 degrees, the direction finding error of 6 degrees exists, the direction finding result of the combined vibration velocity V c is 62 degrees, the direction finding error of 2 degrees exists, and the observation direction finding error of the combined vibration velocity V c is smaller than the fixed direction finding error.
According to the target detection technical scheme, the Toeplitz reconstruction technology applied to the processing of the linear array receiving signal cross covariance matrix is combined with the modal domain transformation technology of the circular array, the circular array is transformed into the virtual linear array, so that the Toeplitz reconstruction technology is indirectly applied to the processing of the circular array receiving signal cross covariance matrix, and then the vector circular array sound pressure vibration velocity combined processing technology is combined, so that the noise suppression capability of an algorithm is further enhanced, and the weak target detection capability is improved.
The background ratio of the weak target is defined as the difference value obtained by subtracting the average background power from the output power of the weak target on the MUSIC spatial spectrum, and the difference value is used for describing the detection capability of the algorithm on the weak target with low signal to noise ratio. The higher the background ratio of the weak target is, the stronger the detection capability of the algorithm to the weak target is, and the easier the azimuth of the weak target with low signal to noise ratio is detected. As can be seen from fig. 5, when the signal-to-noise ratio in the dual-target signal input band is-10 dB and-15 dB respectively, the weak target background ratio of the algorithm provided by the application is improved by about 30dB compared with the array element domain sound pressure MUSIC algorithm and the array element domain vector independent channel MUSIC algorithm, is improved by about 24dB compared with the array element domain (p+v c)*(P+Vc)H MUSIC algorithm), is improved by about 21dB compared with the modal domain (p+v c)*(P+Vc)H MUSIC algorithm), which shows that the weak target detection capability of the algorithm provided by the application is stronger.
According to the target detection technical scheme, the combined vibration velocity V c observation direction phi is combined with the array scanning direction, so that the problem of target azimuth estimation deviation (whether the steps of the application, the drawings, the table and the performance data can be combined or not, and the remarkable improvement of the target azimuth estimation deviation performance compared with the existing method is elaborated) caused by the fact that the observation direction phi is manually selected and the real target azimuth is not consistent is avoided.
As can be seen from the simulation results of fig. 8, the direction finding result for the weak target in the 60 ° direction is 66 ° when the combined vibration velocity V c observation direction is fixed, the direction finding error is 6 °, and the direction finding result for the weak target in the 60 ° direction is 62 ° when the combined vibration velocity V c observation direction matches the array scanning direction, the direction finding error is 2 °, the observation direction matching scanning direction is reduced by 4 ° compared with the observation direction fixing direction finding error, and the 2 ° direction finding error occurring when the combined vibration velocity V c observation direction matches the scanning direction is due to the algorithm itself. This result demonstrates that by matching the combined vibration velocity V c observation direction to the scanning direction, the direction finding error introduced by the fixed observation direction can be avoided.
According to the technical scheme, the Toeplitz matrix reconstruction technology and the vector circular array sound pressure and vibration velocity combined processing technology are combined, two denoising modes of algorithm denoising and array denoising are fully utilized, noise suppression is more sufficient, and the Toeplitz reconstruction technology cannot be directly applied to the circular array, the modal domain transformation technology is applied, so that various technical methods can be organically combined, and the algorithm robustness is improved.
In the existing vector array sound pressure vibration velocity combined processing method, when the combined vibration velocity V c is calculated, the observation direction phi is set manually, if the deviation between the selected observation direction phi and the true target azimuth is larger, a certain azimuth estimation error is generated when the target azimuth estimation is carried out by calculating the sound pressure vibration velocity cross covariance matrix, and the error can be avoided by keeping the observation direction phi consistent with the array scanning vector direction in the technical scheme of the application.
Compared with the prior art, the target detection method in the technical scheme of the application has the advantages of strong noise suppression capability and good weak target detection performance. In particular, the spatial spectral background level is lower and the spatial resolution is higher at low signal-to-noise ratios. The space background level is low because the Toeplitz matrix reconstruction technology and the modal domain transformation technology are organically combined, the method can be fully applied to an array of which the array manifold does not meet the Van der Monte matrix form, and partial noise is removed through Toeplitz matrix reconstruction, so that the signal to noise ratio is improved.
In addition, the characteristic of the vector circular array sound pressure vibration velocity combined processing is fully utilized, the characteristic that the signal sound pressure vibration velocity is coherent and the noise sound pressure vibration velocity is uncorrelated is utilized to calculate a cross covariance matrix, isotropic noise is fully restrained, the signal to noise ratio is improved, most of noise is restrained through superposition of the Toeplitz reconstruction and the sound pressure vibration velocity combined processing on the noise restraining effect, therefore the background level of the output space spectrum is very low, the weak target is difficult to submerge in the background noise, and the detection capability of the weak target is further improved.
The spectrum peak widening resolution of the conventional MUSIC algorithm is reduced under the condition of low signal-to-noise ratio because the orthogonality of a noise subspace and an array scanning vector obtained by characteristic decomposition is destroyed, and the technical scheme of the application improves the accuracy of characteristic decomposition by applying a Toeplitz reconstruction technology, thereby guaranteeing the orthogonality of the noise subspace and the array scanning vector, and the MUSIC spatial spectrum peak is narrower and the resolution is higher due to strong orthogonality when the MUSIC spatial spectrum is calculated.
Those of skill in the art will readily appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is implemented as hardware or computer software driven hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The embodiment of the invention can divide the functional modules according to the method example detection device, for example, each functional module can be divided corresponding to each function, and two or more functions can be integrated in one processing module. The integrated modules may be implemented in hardware or in software functional modules. It should be noted that, in the embodiment of the present invention, the division of the modules is schematic, which is merely a logic function division, and other division manners may be implemented in actual implementation.
Although the invention has been described in connection with specific features and embodiments thereof, it will be apparent that various modifications and combinations can be made without departing from the spirit and scope of the invention. Accordingly, the specification and drawings are merely exemplary illustrations of the present invention as defined in the appended claims and are considered to cover any and all modifications, variations, combinations, or equivalents that fall within the scope of the invention. It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (10)
1. A method of object detection for use in a sub-ice environment, comprising:
carrying out sound pressure and vibration velocity combined processing on the received signals of the vector circular array receiver to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point;
Performing modal domain transformation on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning directions to obtain a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver;
According to the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver, toeplitz reconstruction and characteristic decomposition are carried out, and noise subspace signals corresponding to different scanning orientations are obtained;
according to the noise subspace signals, MUSIC spatial spectrums corresponding to the received signals are obtained by using a MUSIC algorithm;
And determining the target azimuth according to the spectral peak of the MUSIC spatial spectrum.
2. The method of claim 1, wherein the step of determining the position of the substrate comprises,
The sound pressure and vibration velocity combined processing for the received signals of the vector circular array receiver comprises the following steps:
Constructing a broadband receiving signal model based on a vector circular array under ice to obtain a sound pressure receiving signal and x, y vibration speed receiving signals;
Combining the vibration speed signal in the x direction and the vibration speed signal in the y direction to obtain a combined vibration speed signal;
performing cross-correlation processing on the combined vibration velocity signal and the sound pressure signal at each frequency point to obtain a sound pressure vibration velocity cross-covariance matrix at each frequency point;
And traversing the combined vibration velocity in the range of 0-360 degrees in an omnidirectional manner to obtain the sound pressure in each scanning direction and the cross covariance matrix of the combined vibration velocity under different frequency points.
3. The method of claim 2, wherein the step of determining the position of the substrate comprises,
The combined vibration speed observation direction phi is consistent with the scanning direction of the vector circular array.
4. The method of claim 1, wherein the step of determining the position of the substrate comprises,
The method for performing modal domain transformation on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning orientations into the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver comprises the following steps:
Carrying out Fourier transform processing on the received signals of the vector circular array;
Determining a modal transformation matrix according to the signal after the space domain discrete Fourier transformation;
Based on the modal transformation matrix, carrying out modal domain transformation on the sound pressure vibration velocity cross covariance matrix under each scanning direction of different frequency points, and respectively outputting the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver obtained by the modal domain transformation.
5. The method of claim 1, wherein the step of determining the position of the substrate comprises,
The Toeplitz reconstruction and characteristic decomposition are carried out according to the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver, and the method comprises the following steps:
time-averaging the sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver to obtain the sound pressure vibration velocity cross covariance matrix of each scanning direction under different frequencies;
Sequentially carrying out frequency averaging on the cross covariance matrixes of different frequency points under each scanning azimuth to obtain a frequency-averaged cross covariance matrix;
performing Toeplitz reconstruction on the cross covariance matrix after frequency averaging;
And performing feature decomposition on the cross covariance matrix obtained by Toeplitz reconstruction to obtain noise subspaces corresponding to different scanning orientations.
6. The method of claim 1, wherein the step of determining the position of the substrate comprises,
The step of obtaining the MUSIC spatial spectrum corresponding to the received signal by using the MUSIC algorithm according to the noise subspace signal comprises the following steps:
Obtaining the output power of the MUSIC algorithm according to the noise subspaces under different scanning orientations;
And normalizing the MUSIC output power of all scanning directions and taking the logarithm to obtain the MUSIC spatial spectrum.
7. An object detection device for use in an ice environment, comprising:
The joint processing unit is configured to perform sound pressure and vibration velocity joint processing on the received signals of the vector circular array receiver to obtain a sound pressure and vibration velocity cross covariance matrix under each frequency point;
The conversion unit is configured to perform modal domain conversion on the sound pressure vibration velocity cross covariance matrix of each frequency point under different scanning directions to obtain a sound pressure vibration velocity cross covariance matrix of the virtual linear array receiver;
The reconstruction unit is configured to reconstruct Toeplitz and decompose features according to the sound pressure and vibration velocity cross covariance matrix of the virtual linear array receiver to obtain noise subspace signals corresponding to different scanning orientations;
The MUSIC processing unit is configured to obtain a MUSIC spatial spectrum corresponding to the received signal by utilizing a MUSIC algorithm according to the noise subspace signal;
And the azimuth identifying unit is configured to determine a target azimuth according to the spectral peak of the MUSIC spatial spectrum.
8. An object detection device for use in a sub-ice environment, comprising: a vector circular array receiver for receiving signals and an object detection device according to claim 5.
9. A computer storage medium having instructions stored therein which, when executed, implement the object detection method for an under-ice environment of any one of claims 1 to 6.
10. A computing device comprising a processor and a communication interface coupled to the processor; the processor is configured to execute a computer program or instructions to implement the object detection method for an under-ice environment according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410188985.9A CN117949894A (en) | 2024-02-20 | 2024-02-20 | Target detection method, device, medium and equipment for ice environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410188985.9A CN117949894A (en) | 2024-02-20 | 2024-02-20 | Target detection method, device, medium and equipment for ice environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117949894A true CN117949894A (en) | 2024-04-30 |
Family
ID=90796460
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410188985.9A Pending CN117949894A (en) | 2024-02-20 | 2024-02-20 | Target detection method, device, medium and equipment for ice environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117949894A (en) |
-
2024
- 2024-02-20 CN CN202410188985.9A patent/CN117949894A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112269164B (en) | Weak target positioning method based on interference structure matching processing under deep sea reliable acoustic path | |
| CN111198374B (en) | Doppler sensitive signal moving target underwater sound detection method based on space-time-frequency joint interference suppression | |
| WO2009114578A1 (en) | Autonomous sonar system and method | |
| CN108872970B (en) | Grating lobe discrimination method suitable for general equidistant sparse array single-frequency signal beam forming | |
| Sun et al. | Improving the performance of a vector sensor line array by deconvolution | |
| CN104714235A (en) | Ranging method and system for double low-frequency vector hydrophone arrays | |
| CN115236594B (en) | Under-ice sound source positioning method suitable for polar impulse noise environment | |
| CN110673119A (en) | Compressed sensing-based non-regularization orientation estimation method and system | |
| US6424596B1 (en) | Method and apparatus for reducing noise from near ocean surface sources | |
| CN113671473B (en) | Combined matching field positioning method and system based on environment constraint and Riemann distance | |
| CN112285639B (en) | Wideband signal azimuth estimation method based on cross sound pressure array | |
| CN111175727A (en) | Method for estimating orientation of broadband signal based on conditional wave number spectral density | |
| CN115656994B (en) | Real-time calibration method for bistatic active detection towing array shape | |
| CN117949894A (en) | Target detection method, device, medium and equipment for ice environment | |
| CN114563760B (en) | Second-order super-beam forming method, equipment and medium based on SCA array | |
| CN113625222B (en) | Broadband signal high-resolution direction estimation method based on time delay difference norm weighting | |
| CN111722178B (en) | Far-field narrow-band signal incoming wave direction estimation method based on numerical solution of directivity model | |
| CN113109760A (en) | Multi-line spectrum combined DOA estimation and clustering method and system based on group sparsity | |
| CN112924926A (en) | DOA estimation method under multi-path channel | |
| CN112305497A (en) | Near-field microphone array DOA estimation direction finding fuzzy elimination method | |
| Oudompheng et al. | Passive synthetic aperture array to improve noise mapping of a moving ship | |
| Rosenberger | Passive localization | |
| CN112649787B (en) | Target azimuth estimation method based on low-frequency circular array | |
| Wilmut et al. | An efficient target tracking algorithm for matched field processing | |
| Liang et al. | Time-space transform: a novel signal processing approach for an acoustic vector-sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |