CN207663047U - Towing line array array shape estimation device based on single near field correction source - Google Patents

Abstract

The utility model belongs to the field of array signal processing, and aims to realize accurate orientation control of a correction source, thereby improving the accuracy of formation estimation. For this reason, the technical scheme that the utility model adopts is, based on the towed linear array formation estimation device of single near-field correction source, by towing ship, near-field correction source, towed line array, array element, signal acquisition module, signal It consists of a transmission module and a signal processing module; the near-field correction source is installed at the head or tail of the tugboat, and the position coordinates of the correction source are determined by the position sensor, and the array element receives the acoustic signal transmitted through the underwater acoustic channel and converts it into The electrical signal is collected by the signal acquisition module, the analog signal is converted into a digital signal, and the data of each array element is packaged and encoded in the signal transmission module, and transmitted to the signal processing module. The utility model is mainly used in array signal processing occasions.

Description

Towed Line Array Shape Estimation Device Based on Single Near-field Correction Source

technical field

The utility model belongs to the field of array signal processing, in particular to a drag line array shape estimation method and device based on a single near-field correction source.

Background technique

Towed line array sonar has been widely used in marine monitoring and strategic early warning, etc., but due to the change of hull speed, sudden turning, disturbance of ocean currents, etc., the towed array bends, and most array signal processing algorithms such as multiple signal classification ( Multiple Signal Classification, hereinafter referred to as MUSIC), subspace fitting, etc., all realize positioning on the premise that the formation is accurately known, so the formation must be corrected before estimating the azimuth of the sound source. The current active array estimation methods all assume that the auxiliary correction source is a far-field source, and the aperture of the towed line array sonar system is generally large, and it is necessary to place a correction source with an accurate and known azimuth far away from the ship and other detection platforms. Due to the complex sea conditions at sea, it is difficult to ensure the precise orientation of the far-field correction source, which makes the corrected formation have a large error compared with the actual formation, which greatly limits the practical application. In order to overcome the problems faced by using the far-field correction source, the utility model proposes a high-precision towed array formation estimation method and device using a single near-field source. By placing the near-field correction source on the tugboat, the precise azimuth control of the correction source can be realized, thereby improving the accuracy of formation estimation.

Contents of the invention

In order to overcome the deficiencies of the prior art, the utility model aims to propose a high-precision towed line array shape estimation method and device using a single near-field correction source. By placing the near-field correction source on the tugboat, the precise azimuth control of the correction source is realized, thereby improving the accuracy of formation estimation. For this reason, the technical scheme that the utility model adopts is, based on the towed linear array formation estimation device of single near-field correction source, by towing ship, near-field correction source, towed line array, array element, signal acquisition module, signal It consists of a transmission module and a signal processing module; the near-field correction source is installed at the head or tail of the tugboat, and the position coordinates of the correction source are determined by the position sensor, and the array element receives the acoustic signal transmitted through the underwater acoustic channel and converts it into The electrical signal is collected by the signal acquisition module, the analog signal is converted into a digital signal, and the data of each array element is packaged and encoded in the signal transmission module, and transmitted to the signal processing module.

Features and beneficial effects of the utility model are:

The utility model can effectively overcome the shortcomings of poor practicability and large amount of calculation that the existing formation estimation method mostly relies on the far-field correction source, through the same characteristics of the signal subspace and the subspace stretched by the column vectors of the array flow matrix, It can solve the phase difference between different array elements and the reference array element, the calculation method is simple, and the calculation amount is reduced; a single near-field correction source is used for array estimation. Placing the near-field correction source on the towing boat can precisely control the orientation of the correction source, which is in line with the actual application situation and improves the estimation accuracy; and the utility model only uses one near-field source to reduce the error caused by the inaccurate measurement of the correction source position, which can Further reduce the amount of computation.

Description of drawings:

Other purposes and aspects of the present utility model will become clear from the following detailed description with reference to the accompanying drawings. In the accompanying drawings:

Fig. 1 shows a block diagram of the general scheme of the towed line array shape estimation system of the present invention.

Fig. 2 shows a schematic diagram of the formation estimation model based on a single near-field correction source of the present invention.

Fig. 3 shows a block diagram of the formation estimation method based on a single near-field correction source of the present invention.

Fig. 4 shows the average root mean square error of the array element position coordinates under different signal-to-noise ratios SNR in the utility model and CRLB values.

Fig. 5 shows the estimation results of 5 different formations using the method of the utility model.

Fig. 6 shows the DOA estimation result diagram of the MUSIC algorithm before and after correcting the formation using the method proposed by the utility model.

In Fig. 1: 1 is the towing ship; 2 is the near-field correction source; 3 is the underwater acoustic signal emitted by the correction source for correcting the formation; 4 is the towed line array; 5 is the right angle established with the reference array element as the origin Coordinate system; 6 is the array element; 7 is the signal acquisition module; 8 is the signal transmission module; 9 is the signal processing module; 10 is the sea level.

In Fig. 2: 11 is a uniform linear array; 12 is a single near-field correction source; 13 is an array element; 14 is a reference array element; 15 is a rectangular coordinate system established with the reference array element as the origin.

In Fig. 3: 16 is to set the number M of array elements; 17 is to collect acoustic signals for each array element and transmit them to the control center for processing module; 18 is to solve the covariance matrix R of the received data; 19 is to characterize the covariance matrix R Value decomposition; 20 is to solve the phase difference of different array elements from the eigenvector phase; 21 is to solve the phase difference module of receiving acoustic signals of different array elements; 22 is to obtain the position information of the near-field correction source and the reference array element by position sensors such as GPS ; 23 is to solve the distance difference between adjacent array elements; 24 is to solve the distance from the correction source to the reference array element; 25 is to iteratively solve the distance from each array element to the reference array element; Coordinate system; 27 is to solve the polar angle of each array element 28 is to solve the position coordinates (x, y) of each array element; 29 is an array estimation module based on a single near-field correction source.

Detailed ways

The purpose of this utility model is to overcome the disadvantages of poor practicability of the existing formation estimation methods that mostly rely on far-field correction sources, and provide a method for estimating the shape of a towed line array using a single near-field correction source. On the premise that the distance between the array elements is known, by placing the near-field correction source on the towing ship, the azimuth of the correction source can be precisely controlled, which conforms to the actual application situation and improves the estimation accuracy; and the method of the utility model only uses one near-field correction source, The number of variable parameters is reduced and the amount of calculation is reduced.

The first step is to determine the overall scheme of the towed line array shape estimation system.

The block diagram of the overall scheme of the towed line array shape estimation system is shown in Figure 1. It mainly consists of a towed ship 1, a near-field correction source 2, a towed line array 4, an array element 6, a signal acquisition module 7, a signal transmission module 8 and The signal processing module 9 is composed.

The first array element of the array element 6 is set as the reference array element, its position is measured by position sensors such as GPS, and the rectangular coordinate system 5 is established with the reference array element as the origin. The near-field correction source 2 is installed on the head or tail of the tugboat 1 to ensure the precise control of the position of the correction source, and the position coordinates of the correction source are determined by the position sensor, which is used to emit the designed underwater sound for correcting the formation Signal 3, under the premise that the distance between the array elements of the line array is known, the array element 6 receives the acoustic signal transmitted through the underwater acoustic channel, and converts it into an electrical signal, and collects the electrical signal through the signal acquisition module 7, and converts the analog signal Converted into digital signals, each array element data is packaged and encoded in the signal transmission module 8, and transmitted to the signal processing module 9, and the position coordinates of each array element are obtained through the processing of the formation estimation algorithm. The above modules can rely on the digital Signal processing system (Digital Signal Process, hereinafter referred to as DSP), Field-Programmable Gate Array (Field-Programmable Gate Array, hereinafter referred to as FPGA), RISC microprocessor (Acorn RISC Machine, hereinafter referred to as ARM) and other hardware platforms.

The second step is to design a method for solving the phase difference of data received by different array elements.

As shown in Figure 2, the uniform linear array 11 is bent due to the influence of ocean currents or ship speed changes. Assuming that the reflection of the correction acoustic signal and the change of the array elements in the depth direction can be ignored, a single near-field correction source 12 emits sound The signal is received by M hydrophones 13 (array elements), and the array received data vector can be expressed as,

X(t)=AS(t)+N(t) (1)

Where X(t) is the M×1-dimensional snapshot data vector of the array, A is the M×1-dimensional array flow pattern matrix, S(t) is the spatial signal vector, and N(t) is the M×1-dimensional noise data of the array vector. A rectangular coordinate system 15 is established with the reference array element 14 as the origin, (xi _, y _i )(i=2,3,...,M) is the position coordinate of the i-th array element, and θ is the DOA of the correction source. Fig. 3 is a block diagram of a formation estimation method based on a single near-field correction source. For solving the phase difference part 21 of data received by different array elements, first set the number of array elements M 16, and all array elements collect the acoustic signals emitted by the correction source and send them to the control center for processing 17. This part can rely on DSP, FPGA, ARM Wait for the hardware platform to be realized, then solve the covariance matrix R 18 of the received data vector, and perform eigenvalue decomposition on R 19, according to the theory of array signal processing, the signal subspace and the subspace formed by the column vectors of the array flow matrix A are the same , so the phase difference 20 of different array elements can be solved according to the phase of the eigenvector, which can be used to further estimate the position coordinates of the array elements.

The third step is to design a formation estimation model based on a single near-field correction source.

As shown in FIG. 2 , assuming that there is a straight line between adjacent array elements, the distance is d, the distance from the near-field correction source S ₁ (x _a , y _a ) to the reference array element 14 is r ₁ , and the DOA is θ. ( _xi , y _i ) is the position coordinate of the i-th array element, and _ri is the distance from the calibration source S ₁ to the i-th array element. The formation estimation module 29 based on a single near-field source as shown in FIG. The phase difference △φ _i between different array elements solved in the second step can obtain the distance difference 23 between two adjacent array elements and the reference array element, expressed as follows,

r _i ＝r _i-1 -△φ _i ·λ/(2·π), i=2,3,...,M (2)

According to the initially calculated distance r ₁ 24 from the near-field correction source to the reference array element, the distance 25 from each array element to the reference array element can be iteratively calculated. Because the acoustic signal emitted by the near-field correction source is a spherical wave, the polar coordinate system 26 is established with the near-field correction source _S1 as the origin, then the polar angle of the i-th array element can be expressed as,

It can be seen from Fig. 2 that the rectangular coordinates ( _xi , y _i )28 of the i-th array element can be calculated by the following formula,

In the fourth step, the Cramér-Rao low bound (CRLB for short) of the proposed formation estimation method based on a single near-field correction source is derived. By comparing different Signal-Noise Ratio (SNR, The root mean square error (Root-Mean-Square Error, RMSE for short) and the CRLB value of the position coordinates of the array element under the SNR, and the root mean square error RMSE decreases with the increase of the SNR, and is consistent with The conclusion that there is little difference between CRLB and CRLB verifies the correctness of this method.

The technical scheme of the utility model is that the first array element of the towed linear array is set as a reference array element, its position is measured by a position sensor such as GPS, and a rectangular coordinate system is established with the reference array element as the origin, and the nearly The field correction source is installed at the head or tail of the tugboat, and the position coordinates of the correction source are determined by the position sensor, which is used to transmit the designed underwater acoustic signal for correcting the formation, and the array element receives the sound transmitted through the underwater acoustic channel. signal, and convert it into an electrical signal, and transmit it to the control center through the signal acquisition module and the transmission module. This part can be realized by relying on hardware platforms such as DSP, FPGA, and ARM. Based on the phase difference of the correction source and the geometric relationship between the array elements, an array estimation model based on a single near-field correction source is established to solve the position coordinates of each array element.

The utility model will be further described in detail below in conjunction with the accompanying drawings and specific embodiments of the utility model.

The first step is to place the near-field correction source and set the relevant parameters of the emitted acoustic signal.

According to the near-field source distance condition of the array 0.62·(D ³ /λ) ^1/2 <r≤2D ² /λ, where D is the array aperture, λ is the central wavelength of the signal source; r is the distance from the signal source to the reference array element Distance, fix the near-field calibration source at a proper position at the stern of the towed ship, measure and transform the rectangular coordinates of the near-field calibration source through GPS and other position sensors as (0,150), then the distance from the near-field calibration source to the reference array element is 150m, reaching The azimuth angle DOA is 0; the emission sound signal is a sinusoidal signal, and the center frequency is 125Hz; the uniform line array is selected in the experiment, the number of array elements M is 25, and the array element spacing d is 3m; the sampling frequency of the signal acquisition module is 1kHz; The number of repeated experiments was 100 times to avoid the chance of experimental results.

The second step is to solve the phase difference of the acoustic signals received by different array elements.

The array element receives the acoustic signal transmitted through the underwater acoustic channel, converts it into an electrical signal, and transmits it to the control center for processing through the signal acquisition module and the transmission module. Bending, assuming that the reflection of the correction acoustic signal and the change of the array element in the depth direction can be ignored, the acoustic signal emitted by the placed near-field correction source 12 is received by 25 hydrophones 13 (array elements), and in the 25×1-dimensional The array flow pattern matrix A contains the related information of the array shape, which can be expressed as,

Where φ _k is the phase from the near-field correction source to the kth array element, k=1,2,,25. Expressed as,

Where r is the distance from the near-field calibration source to the reference array element 14, which is 150m, λ is the center wavelength of the calibration source, which is 12m, and (x _k ,y _k )(k=1,2,...,M) is The position coordinate of the kth array element, θ is the DOA from the calibration source to the reference array element, which is 0. Further, solve the covariance R 18 of the received data vector, and perform eigenvalue decomposition 19 on R, expressed as follows,

where Σ _s is a diagonal matrix composed of eigenvalues ξ _k , and e _k is the corresponding eigenvector. According to the theory of array signal processing, the signal subspace is the same as the subspace formed by the column vectors of the array flow matrix A. According to the eigenvector Solve the phase difference of different array elements by 20, which is further expressed as, phase φ _k = Arg(e _k ), then the phase difference from the i-th array element to the k-th array element is,

φ _k -φ _i =Arg(e _k )-Arg(e _i )i=2,3,...,M; k=1,2,...,M (8)

From this, the phase difference between different array elements can be obtained, which is used to estimate the position coordinates of the array elements.

The third step is to solve the array element position coordinates according to the formation estimation model based on a single near-field source.

According to the phase difference between different array elements solved in the second step, the distance difference 23 between two adjacent array elements can be obtained, expressed as follows,

△d _i = △φ _i ·λ/(2·π) (9)

According to the initially calculated distance r ₁ 24 from the near-field correction source to the reference array element and the distance difference between adjacent array elements and the reference array element, the distance 25 from each array element to the reference array element can be calculated iteratively, expressed as follows,

r ₁ =150m (10)

r _i ＝r _i-1 -△φ _i ·λ/(2·π) (11)

Because the acoustic signal emitted by the near-field correction source is a spherical wave, the polar coordinate system 26 is established with the near-field correction source _S1 as the origin, then the polar angle of the i-th array element can be expressed as,

It can be seen from Fig. 2 that the rectangular coordinates (xi _, y _i ) 28 of the i-th array element can be expressed as,

The 4th step is to derive and analyze the error and CRLB about array element coordinates in the utility model method

Define the phase vector of all array elements relative to the reference array element to receive data as Φ=[φ ₂ ,φ ₃ ,...,φ _M ], then the CRLB about Φ can be expressed by the reciprocal of the Fisher information matrix J,

CRLB(Φ)＝J ^-1 (14)

Among them, ρ(X|Φ) is the probability density matrix of the received data matrix X with respect to the phase vector Φ. Since the received data matrix X obeys the complex Gaussian distribution with zero mean value and the data sampling points are relatively independent, the correlation between the Fisher information matrix J and the received data The variance matrix R is expressed as,

Where N is the number of sampling points, A(Φ) is the array flow matrix, is the signal power of the near-field correction source, is the noise signal power. The first derivative of the covariance matrix R with respect to the phase φ _i of the i-th array element and the inverse of R are expressed as,

Among them, M is the number of array elements, V _i is a (M-1)×(M-1) dimensional matrix, its position on the diagonal is 1, and the rest are 0, and I is (M -1)×(M-1) dimensional matrix, the signal-to-noise ratio SNR can be expressed as So CRLB(Φ) is expressed as follows,

Where 1 is a (M-1)×1-dimensional vector with all elements being 1, η=(M+1/SNR)/(MN·SNR),

Define the x-coordinate vector of the array element as x=[x ₂ (Φ),x ₃ (Φ),...,x _M (Φ)] ^T , then CRLB(x) can be obtained by transforming the vector parameters of CRLB(Φ) , the relationship between the two is expressed as follows,

So the CRLB(x _i ) and CRLB(y _i ) of the i-th array element coordinates are expressed as,

according to Deriving CRLB( _xi ,y _i ) as,

in,

The average root mean square error value of all elements of the towed line array The solution is as follows, M is the number of array elements, and P is the number of Monte Carlo experiments. By comparing the CRLB of the element coordinates with The performance of the proposed formation estimation method can be further analyzed.

As shown in Figure 4, the average root mean square error of the array element position coordinates under different signal-to-noise ratios SNR and CRLB values, as can be seen from Figure 4, the average root mean square error of the array element coordinates It decreases with the increase of SNR, and has little difference with CRLB, which verifies the correctness of the method of the utility model.

Fig. 5 is the estimation result of 5 different formations using the method of the present utility model. It can be seen that the formation is corrected by the method of the present utility model, and the degree of fitting between the estimated formation and the actual formation is relatively high, and the formation can be accurately realized. estimate.

Fig. 6 is a DOA estimation result diagram of the MUSIC algorithm before and after correcting the formation using the method proposed by the utility model, and three far-field source azimuth angles DOA to be estimated are selected as -30, 30, and 80. It can be seen from the figure that the DOA estimation is blurred before the formation correction, and the DOA estimation results after correction are -29.99, 29.95, 79.98, which are not much different from the actual azimuth angle, which effectively improves the DOA estimation performance of the MUSIC algorithm, proving that The effectiveness of the utility model method has been proved.

Claims (1)

1. A towed line array shape estimation device based on a single near-field correction source, characterized in that it consists of a towed ship, a near-field correction source, a towed line array, an array element, a signal acquisition module, a signal transmission module and a signal The processing module consists of the near-field correction source installed on the head or tail of the towing ship, and the position coordinates of the correction source are determined by the position sensor. The array element receives the acoustic signal transmitted through the underwater acoustic channel and converts it into an electrical signal. The signal acquisition module collects the electrical signal, converts the analog signal into a digital signal, packs and encodes each element data in the signal transmission module, and transmits it to the signal processing module.