A Three-Point Ranging Method Based on Secondary Denoising Time Delay Matching

Abstract

The present invention relates to a three-point ranging method based on secondary denoising time-delay matching. By constructing an array signal based on a secondary correlation function and a steering vector based on frequency and time-delay, time-delay search is performed in a preset manner to obtain the time-delay difference between any one of the three array elements and the other two array elements, thereby obtaining the distance to the target. The beneficial effect of the present invention is that in the case of low signal-to-noise ratio, the secondary correlation function can effectively suppress noise, and the beamforming time-delay estimation can greatly improve the time-delay estimation accuracy without upsampling, thus improving the ranging accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of computing, reckoning or counting, and particularly relates to a three-point ranging method based on quadratic denoising time-delay matching in the field of sonar signal processing. Background Art

[0002] Ternary array passive positioning is a widely used technology in underwater acoustic passive positioning. Because it requires little prior knowledge and is simple and practical, it has become one of the most commonly used passive ranging sonars. The working principle of the ternary array is to utilize the change of the spherical wave front curvature, and by measuring the radiated noise signals received by each element from the target, estimate the time-delay difference between the elements and then calculate the target azimuth and distance.

[0003] In the technology of ternary array passive positioning, the accuracy of the time-delay difference directly affects the performance of passive ranging. Therefore, how to improve the estimation accuracy of the time-delay difference between different elements has also become a research hotspot. Currently, commonly used time-delay difference estimation methods include ordinary cross-correlation, generalized cross-correlation, quadratic correlation, etc. These methods have good time-delay estimation effects under high signal-to-noise ratios. However, in most actual environments, the target is far away and the signal-to-noise ratio is low. Therefore, the above commonly used methods are not very ideal under low signal-to-noise ratio conditions. Summary of the Invention

[0004] The present invention solves the problems existing in the prior art and provides an optimized three-point ranging method based on quadratic denoising time-delay matching.

[0005] The technical solution adopted by the present invention is a three-point ranging method based on quadratic denoising time-delay matching, which constructs an array signal based on a quadratic correlation function and a steering vector based on frequency and time-delay, and performs time-delay search in a preset manner to obtain the time-delay difference between any one element of the three-point array and the other two elements, so as to obtain the distance of the target.

[0006] Preferably, the preset manner is a beamforming method.

[0007] Preferably, the method includes the following steps:

[0008] Step 1: Obtain three-point elements, namely the first element, the second element and the third element;

[0009] Step 2: Construct the autocorrelation function R 11 and the cross-correlation function R 12 of the output signals x1 and x2 of the first element and the second element; construct the autocorrelation function R 22 and the cross-correlation function R 23 of the output signals x2 and x3 of the second element and the third element;

[0010] Step 3: Take R 11and R 12 and R 22 and R 23 are respectively defined as new input signals, decomposed into multiple narrowbands through Fourier transform, and the frequency-domain cross-spectrum matrix R is set for each group of narrowbands respectively 1112 (f k ), R 2223 (f k ); construct the steering vector V;

[0011] Step 4: Set a reasonable time-delay search vector according to the limitations of the search distance and calculation time, calculate the time-delay search spectrum for each group of narrowbands respectively; aggregate the spatial spectra of all narrowbands to obtain the corresponding time-delay spatial spectrum of the broadband;

[0012] Step 5: Search for the peak position of the time-delay spectrum, and determine the time-delay differences between the first and second array elements and between the second and third array elements respectively;

[0013] Step 6: Based on the two time-delay differences, obtain the target distance.

[0014] Preferably, in the said Step 3, the signal vector X 11 and R 12 for constructing R 12 =[R 11 ; R 12 , and the signal vector X 22 and R 23 for constructing R 23 =[R 22 ; R 23 ;

[0015] The frequency-domain cross-spectrum matrices set for each group of narrowbands are respectively R 1112 (f k ) = E|X 11 (f k )·X 12 H (f k )|, R 2223 (f k ) = E|X 22 (f k )·X 23 H (f k )|, where f k is the frequency point within any narrowband.

[0016] Preferably, in the said Step 3, the steering vector V = [a(f k , t)], f k is the frequency point within any narrowband, and t is the narrowband time delay of the array in the f k frequency band.

[0017] Preferably, in the step 4, the delay search spectra corresponding to each group of narrow bands are respectively P 1112 (f k ,t) = a(f k ,t) H R 1112 (f k )a(f k ,t) and P 2223 (f k ,t) = a(f k ,t) H R 2223 (f k )a(f k ,t).

[0018] Preferably, the delay spatial spectrum corresponding to the corresponding wide band is and

[0019] Preferably, in the step 5, the delay difference τ 12 between the first array element and the second array element and the delay difference τ 23 between the second array element and the third array element are determined;

[0020] In the step 6, the target distance is r, where c is the speed of sound and d is the three-element array spacing.

[0021] The present invention relates to an optimized three-point ranging method based on quadratic denoising delay matching, which constructs an array signal based on a quadratic correlation function and a steering vector based on frequency and delay, performs delay search in a preset manner, obtains the delay difference between any one of the three array elements and the other two array elements, and obtains the distance of the target.

[0022] The beneficial effect of the present invention is that: in the case of low signal-to-noise ratio, the quadratic correlation function can effectively suppress noise, and the beamforming delay estimation can greatly improve the delay estimation accuracy without upsampling, thereby improving the ranging accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flowchart of the present invention;

[0024] Figure 2 is a comparison of the delay estimation results in the present invention and the conventional cross-correlation delay estimation;

[0025] Figure 3 is a comparison of the ranging results in the present invention and the ranging results relying on the conventional cross-correlation delay estimation under different signal-to-noise ratio conditions. DETAILED DESCRIPTION OF THE INVENTION

[0026] The present invention will be further described in detail below in conjunction with embodiments, but the protection scope of the present invention is not limited thereto.

[0027] The present invention relates to a three-point ranging method based on secondary denoising time-delay matching, constructs array signals based on a secondary correlation function and steering vectors based on frequency and time-delay, performs time-delay search in a preset manner, obtains the time-delay differences between any one of the three array elements and the other two array elements, and obtains the distance of the target.

[0028] The preset manner is a beamforming method.

[0029] In the present invention, in the case of low signal-to-noise ratio, the secondary correlation function can effectively suppress noise, and the beamforming time-delay estimation can greatly improve the time-delay estimation accuracy without upsampling, and the ranging accuracy is high.

[0030] In the present invention, specifically, the steering vector V is a function of frequency f and time-delay t.

[0031] As Figure 1 shown, the method includes the following steps:

[0032] Step 1: Obtain three array elements, namely the first array element, the second array element, and the third array element;

[0033] Step 2: Construct the autocorrelation function R 11 and the cross-correlation function R 12 of the output signals x1 and x2 of the first array element and the second array element; construct the autocorrelation function R 22 and the cross-correlation function R 23 of the output signals x2 and x3 of the second array element and the third array element;

[0034] In the present invention, constructing the autocorrelation function R 11 and the cross-correlation function R 12 of x1 and x2, and the autocorrelation function R 22 and the cross-correlation function R 23 of x2 and x3 are easy for those skilled in the art to understand, and those skilled in the art can set them according to needs.

[0035] Step 3: Define R 11 and R 12 as well as R 22 and R 23 as new input signals respectively, decompose them into multiple narrow bands through Fourier transform respectively, and set the frequency-domain cross-spectrum matrices R 1112 (f k ) and R 2223 (f k ) corresponding to each group of narrow bands respectively; construct the steering vector V;

[0036] In step 3, construct R 11 and R 12 signal vector X 12 = [R 11 ; R 12 , construct R 22 and R 23 signal vector X 23 = [R 22 ; R 23 ;

[0037] The frequency-domain cross-spectrum matrices corresponding to each group of narrowbands are respectively R 1112 (f k ) = E|X 11 (f k )·X 12 H (f k )|, R 2223 (f k ) = E|X 22 (f k )·X 23 H (f k )|, where f k is the frequency point within any narrowband.

[0038] In step 3, the steering vector V = [a(f k , t)], f k is the frequency point within any narrowband, and t is the narrowband delay of the array at the f k frequency band.

[0039] In the present invention, constructing X 12 = [R 11 ; R 12 and X 23 = [R 22 ; R 23 is easy for those skilled in the art to understand, and those skilled in the art can set it according to their needs.

[0040] Step 4: Set a reasonable delay search vector according to the limitations of the search distance and calculation time, calculate the delay search spectrum for each group of narrowbands respectively; aggregate the spatial spectra of all narrowbands to obtain the corresponding wideband delay spatial spectrum;

[0041] In step 4, the delay search spectra corresponding to each group of narrowbands are respectively P 1112 (f k , t) = a(f k , t) H R 1112 (f k )a(f k , t) and P2223 (f k , t) = a(f k , t) H R 2223 (f k ) a(f k , t).

[0042] The corresponding delay spectrum of the broadband is and

[0043] Step 5: Search for the peak positions of the delay spectrum, and determine the delay differences between the first and second array elements and between the second and third array elements respectively;

[0044] In the said Step 5, determine the delay difference τ 12 between the first and second array elements and the delay difference τ 23 between the second and third array elements;

[0045] Step 6: Based on the two delay differences, obtain the target distance.

[0046] In the said Step 6, the target distance is r, where c is the speed of sound, d is the spacing of the three - element array, and the three - point ranging here is defaulted to an equally - spaced array.

[0047] In the present invention, by searching for the peak positions of the delay spectra of P 12 and P 23 , the delay difference τ 12 between the first and second array elements and the delay difference τ 23 between the second and third array elements can be determined, and then the target distance can be obtained through calculation based on the delay differences.

[0048] As Figure 2 shown, it is a comparison between the delay estimation result in the present invention and the conventional cross - correlation delay estimation. It can be seen from the figure that the height of the sidelobe of the secondary denoising delay matching is significantly lower than that of the conventional cross - correlation, which can effectively reduce the delay estimation error;

[0049] As Figure 3 shown, it is a comparison between the ranging result in the present invention and the ranging result relying on the conventional cross - correlation delay estimation under different signal - to - noise ratio conditions. It can be seen that the ranging error based on the secondary denoising delay matching is significantly smaller than that of the conventional cross - correlation.

Claims

1. A three - point ranging method based on quadratic denoising time - delay matching, characterized in that: Construct array signals based on the quadratic correlation function and steering vectors based on frequency and time delay, perform time delay search in a preset manner, obtain the time delay differences between any one of the three array elements and the other two array elements, and obtain the distance to the target; The method includes the following steps: Step 1: Obtain three array elements, namely the first array element, the second array element, and the third array element; Step 2: Construct the autocorrelation function \(R\) of the output signals \(x1\) and \(x2\) of the first and second array elements 11 and the cross-correlation function \(R\) 12 ; Construct the autocorrelation function \(R\) of the output signals \(x2\) and \(x3\) of the second and third array elements 22 and the cross-correlation function \(R\) 23 ; Step 3: Define R 11 and R 12 as well as R 22 and R 23 as new input signals respectively, construct the signal vectors X 11 and R 12 as X 12 = [R 11 ; R 12 , construct the signal vectors X 22 and R 23 as X 23 = [R 22 ; R 23 ; Decompose them into multiple narrowbands through Fourier transform respectively, and set the frequency-domain cross-spectrum matrices R 1112 (f k ), R 2223 (f k ) corresponding to each group of narrowbands respectively. The frequency-domain cross-spectrum matrices set for each group of narrowbands are , respectively, where f k is the frequency point within any narrowband; construct the driving vector V, V = [a(fk,t)], f k is the frequency point within any narrowband, and t is the narrowband time delay of the array at the f k frequency band; Step 4: Set a reasonable time-delay search vector according to the limitations of the search distance and calculation time, calculate the time-delay search spectrum for each group of narrowbands respectively, and the time-delay search spectra corresponding to each group of narrowbands are and ; After aggregating the spatial spectra of all narrowbands respectively, the corresponding time-delay spatial spectrum of the broadband is obtained, and the corresponding time-delay spatial spectra of the broadband are and ; Step 5: Search for the spectral peak positions of the time delay spatial spectrum, and respectively determine the time delay differences between the first array element and the second array element, and between the second array element and the third array element; Step 6: Based on the two time delay differences, obtain the target distance.

2. The three - point ranging method based on quadratic denoising time - delay matching according to claim 1, characterized in that: The preset manner is the method of beamforming.

3. The three - point ranging method based on quadratic denoising time - delay matching according to claim 1, characterized in that: In the said step 5, determine the time delay difference between the first array element and the second array element , and the time delay difference between the second array element and the third array element ; In step 6, the target distance is r, , where c is the speed of sound and d is the triple array spacing.

Images