CN111257891B - Deconvolution-based MIMO sonar distance sidelobe suppression method - Google Patents
Deconvolution-based MIMO sonar distance sidelobe suppression method Download PDFInfo
- Publication number
- CN111257891B CN111257891B CN202010083253.5A CN202010083253A CN111257891B CN 111257891 B CN111257891 B CN 111257891B CN 202010083253 A CN202010083253 A CN 202010083253A CN 111257891 B CN111257891 B CN 111257891B
- Authority
- CN
- China
- Prior art keywords
- substep
- mimo sonar
- mimo
- sonar
- point spread
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/88—Sonar systems specially adapted for specific applications
- G01S15/89—Sonar systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/52003—Techniques for enhancing spatial resolution of targets
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/523—Details of pulse systems
- G01S7/526—Receivers
- G01S7/527—Extracting wanted echo signals
- G01S7/5273—Extracting wanted echo signals using digital techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/523—Details of pulse systems
- G01S7/526—Receivers
- G01S7/53—Means for transforming coordinates or for evaluating data, e.g. using computers
Abstract
The invention relates to a deconvolution-based MIMO sonar distance sidelobe suppression method, which is characterized in that an optimal point spread function is designed according to orthogonal waveforms used by MIMO sonar, and deconvolution processing is performed on beam output absolute values by the optimal point spread function and a Richardson-Lucy algorithm, so that distance sidelobes in an MIMO sonar imaging result are effectively suppressed. Compared with the traditional MIMO sonar imaging method, the deconvolution-based MIMO sonar range sidelobe suppression method can obviously reduce the range sidelobe in the imaging output.
Description
Technical Field
The invention belongs to the field of sonar imaging, and particularly relates to a deconvolution-based MIMO sonar distance sidelobe suppression method.
Background
In the field of underwater acoustic Imaging, a multiple-input multiple-output (MIMO) Sonar can achieve angular Resolution (Liu XH, Sun C, Zhuo J, and et al, "developing MIMO array for unwater 3-D short-range Imaging," in Proc MTS/IEEE OCEANS ' 12, Hampton Roads, USA, pp.1-7, Oct.2012.Sun C, Liu X H, Zhuo J, and et al, "High-Resolution 2-D Sector-scanning MIMO Sonar with LFpumps," in Proc ' 13MTS/IEEE, Sano MIMO's 13 MIMO SOx and USA) superior to conventional single-input multiple-output (SIMO) Sonar. However, due to imperfect orthogonality between the used transmission waveforms and waveform distortion caused by an underwater acoustic channel, high distance dimension sidelobe interference exists in the imaging result output of the MIMO sonar (Liuxiong, Suntang, Zhujie, etc.. mismatch filtering processing of the MIMO imaging sonar based on the convex optimization method. northwest university proceedings, 2013; 31(3): 367-. Although the use of a longer code sequence can suppress the range-dimension side lobe to some extent, it is easy to cause problems such as an excessive calculation amount and poor waveform robustness. The existing research cannot solve the problem of effectively inhibiting the distance dimension side lobe on the premise of not increasing the length of the coded sequence.
Disclosure of Invention
The invention solves the technical problems that: in order to suppress range dimension sidelobes in an MIMO sonar imaging result, the invention provides a deconvolution-based MIMO sonar range sidelobe suppression method. Aiming at the original imaging output of the MIMO sonar, the method designs a point spread function according to an orthogonal waveform used by the MIMO sonar, and performs deconvolution processing on an absolute value output by each wave beam by using a Richardson-Lucy algorithm and the designed point spread function, so as to achieve the purpose of remarkably reducing the distance side lobe in an imaging result.
The technical scheme of the invention is as follows: a deconvolution-based MIMO sonar range sidelobe suppression method comprises the following steps:
the method comprises the following steps: obtaining original MIMO sonar imaging output by an MIMO sonar imaging method:
step two: designing an optimal point spread function for deconvolution processing from orthogonal waveforms used by the MIMO sonar, comprising the sub-steps of:
the first substep: preserving the main lobe part of the autocorrelation function in the orthogonal multi-phase coding signal used by the MIMO sonar, removing the side lobe part of the autocorrelation function, designing a positive number eta as the side lobe level of the point spread function, and obtaining the expression of the point spread function PSF (t)
Wherein T represents time, T 0 Representing the length of a sub-code in the quadrature polyphase encoded signal;
optimizing the value of eta, and setting the value variable interval of eta, which is contained in 10 -6 To 10 -0.01 Obtaining a plurality of point spread functions PSF (t) under different eta values in the purchased interval;
and a second substep: obtaining a correlation function R (t) according to orthogonal multiphase coded signals used by the MIMO sonar:
wherein R is m,m (t) denotes an autocorrelation function of the mth (M ═ 1,2, …, M) orthogonal polyphase coded signal, R m,i (t) represents a cross-correlation function between the mth orthogonal polyphase coded signal and other orthogonal polyphase coded signals, wherein M represents the number of orthogonal polyphase coded signals used by the MIMO sonar and satisfies that M is more than or equal to 2;
and a third substep: the absolute value of the correlation function r (t) is obtained | r (t) |, where | | represents the absolute value. And (3) taking the absolute value R (t) and the point spread function PSF (t) as the input of the Richardson-Lucy algorithm, and performing deconvolution processing. The Richardson-Lucy algorithm is an iterative algorithm, and the k +1 th iteration result r (k+1) (t) is represented by
Where k represents the number of iterations.
As the iteration progresses, the iteration result r (k) (t) will converge to a unique solution that minimizes the Csiszar distance, corresponding to the expression:
wherein the content of the first and second substances,
r (t) representing solving so that the function in parentheses takes the minimum value, and L () representing the Csiszar distance. Let a (x) and b (x) be two variables, the expression for the Csszar distance is
And a fourth substep: and repeating the substep three to obtain the output of the Richardson-Lucy algorithm under all the eta values. Solving the highest sidelobe level of the outputs of the Richardson-Lucy algorithms, and finding out the eta value corresponding to the lowest value of all the highest sidelobe levels to be eta value 0 . By η 0 Constructing point spread function PSF (t, eta) for MIMO sonar range sidelobe suppression 0 ) The expression is
Step three: and (3) deconvoluting the MIMO sonar original program imaging output obtained in the first step by using the optimal point spread function and the Richardson-Lucy algorithm designed in the second step to reduce the distance side lobe, and the method comprises the following substeps:
the first substep: the input of the Richardson-Lucy algorithm is an optimal point spread function
And the Q-th (Q-1, 2, …, Q) beam output B q Absolute value of (t) | B q (t) |, output being the target distribution S obtained for the kth iteration (k) (t)；
And a second substep: in the iterative process of reducing distance sidelobe, the k +1 iteration result S of the Richardson-Lucy algorithm (k+1) (t) is represented by
As the iteration progresses, S (k) (t) will converge to a unique solution that minimizes the Cssizar distance
Wherein, the first and the second end of the pipe are connected with each other,
represents s (t) for solving so that the function in parentheses takes the minimum value.
And a third substep: repeating the second substep, using the formula point spread function PSF (t, eta) for the absolute values of all beam outputs 0 ) And performing deconvolution processing by a Richardson-Lucy algorithm to obtain imaging output after sidelobe suppression.
The further technical scheme of the invention is as follows: the step one of obtaining the MIMO sonar imaging output specifically includes the following substeps:
the first substep: defining MIMO sonar comprising M transmitting transducers and N receiving hydrophones, wherein the M-element transmitting transducers form an M-element uniform transmitting linear array, and the N receiving hydrophones form an N-element uniform receiving linear array; transmitting transducer spacing d t Spaced from the receiving hydrophone by a distance d r Satisfy the requirement of
d t ＝Nd r ；
And a second substep: the MIMO sonar uses M (M is more than or equal to 2) orthogonal coding signals with the same frequency band, and the M (M is 1,2, … …, M) th signal is s m (t), the expression of which is:
wherein T represents time, L represents the number of subcodes, T 0 For the length of a single sub-code,
for the initial phase, f, of the l-th sub-code in the m-th transmitted signal 0 J represents the imaginary part for the center frequency of the transmitted signal;
and a third substep: let the region to be imaged in the far field be modeled as P discrete points, and the received signal on the nth receiving array element be:
wherein σ p Is the scattering intensity of the p-th scattering point,
the time delay from the m-th transmitting array element to the p-th scattering point,
is the time delay from the p scattering point to the n receiving array element, and n (t) is additive noise;
and a fourth substep: neglecting the matched filtering output of the noise N (t), performing matched filtering on the receiving signals on the N receiving array elements by using copies of M transmitting signals at a receiving end to obtain MN matched filtering outputs, wherein the (M-1) N + N matched filtering outputs are obtained by the matched filtering output of the mth transmitting signal copy on the echo signal on the nth receiving array element and are expressed as
R m,m (t) is the autocorrelation function of the mth transmitted signal, R m,i As a cross-correlation function between the mth transmit signal and the other transmit signals,
and
have the same meaning, but satisfy i ≠ m;
and a fifth substep: performing beam forming processing on the MN matched filter outputs; let the Q (Q is 1,2, …, Q) th beam output be B q (t)
Wherein the content of the first and second substances,
the weighting for the (m-1) N + N matched filter outputs,
for corresponding amplitude weighting, theta q Is the scan angle on the q-th beam;
will be provided with
Obtaining | B by solving absolute value of all beam outputs q (t) and splicing together according to the beam scanning sequence to obtain the original imaging output of the MIMO sonar.
Effects of the invention
The invention has the technical effects that: the basic principle and the implementation scheme of the invention are verified by computer numerical simulation, and the result shows that: the method provided by the invention designs the optimal point spread function according to the orthogonal waveform used by the MIMO sonar, and performs deconvolution processing on the beam output absolute value by using the optimal point spread function and the Richardson-Lucy algorithm, thereby effectively inhibiting the distance side lobe in the MIMO sonar imaging result. Compared with the traditional MIMO sonar imaging method, the MIMO sonar range sidelobe suppression method based on deconvolution provided by the invention can obviously reduce the range sidelobe in imaging output.
Drawings
Fig. 1 is an M-transmit N-receive MIMO sonar array type used in a conventional MIMO sonar imaging method, in which a solid circle represents a transmitting transducer and a solid rectangle represents a receiving hydrophone;
FIG. 2 is a basic process flow of the present invention;
FIG. 3 is a flow chart of the optimal point spread function design according to the method of the present invention;
FIG. 4 is a side lobe suppression process flow of the deconvolution method of the present invention;
fig. 5 shows the design result of the optimal point spread function in the embodiment. Wherein fig. 5(a) shows the side lobe levels obtained at different values of η, the result being obtained by calculating the average of 100 side lobe levels using 100 cycles. FIG. 5(b) is the optimal point spread function calculated from FIG. 5 (a); FIG. 5(c) is a top view of the correlation function obtained for 100 cycles; FIG. 5(d) is a top view of the results obtained by deconvolving the correlation function in 100 cycles using the optimal point spread function and the Richardson-Lucy algorithm of FIG. 5 (b);
fig. 6 shows the original imaging result of MIMO sonar and the result of the deconvolution method proposed by the present invention when the target is a single scattering point target in the implementation example. Fig. 6(a) shows the original imaging result of MIMO sonar, fig. 6(b) shows the result of the deconvolution method of the present invention, and fig. 6(c) shows the distance dimension slice of the two methods at the beam scan angle of 0 °.
Fig. 7 shows the original imaging result of MIMO sonar and the result of the deconvolution method proposed by the present invention when the target is a multi-scattering-point target in the example. Fig. 7(a) shows the original imaging result of MIMO sonar, fig. 7(b) shows the result of the deconvolution method proposed by the present invention, and fig. 7(c) shows the projections of the two methods in the distance dimension.
Detailed Description
Referring to fig. 1-7, the main contents of the present invention are:
1. aiming at the orthogonal waveform used by the MIMO sonar, a corresponding Point Spread Function (PSF) is designed
2. And designing a deconvolution-based MIMO sonar distance sidelobe suppression flow. The procedure uses the designed PSF and Richardson-Lucy algorithm to perform deconvolution processing on the absolute value output by each wave beam in the MIMO sonar imaging result to obtain the imaging result after the range sidelobe suppression
3. The MIMO sonar imaging result and the side lobe suppression method result are provided through computer numerical simulation, and the side lobe suppression method has lower distance side lobes compared with the existing MIMO sonar imaging method, which is shown from the angle of the distance side lobe level in the imaging results of the single scattering point target and the multiple scattering point target.
Technical scheme of the invention
The technical scheme adopted by the invention for solving the existing problems can be divided into the following 3 steps:
1) and obtaining the original imaging output of the MIMO sonar by using the existing MIMO sonar imaging method.
2) And designing a point spread function according to the orthogonal waveform used by the MIMO sonar.
3) And carrying out deconvolution processing on the absolute value output by each beam by utilizing a Richardson-Lucy algorithm and a designed point spread function to obtain an imaging result after distance sidelobe suppression.
Each step of the present invention is described in detail below:
step 1) mainly relates to the use of the existing MIMO sonar imaging method to obtain the original imaging output, and the related theories and specific contents are as follows:
the MIMO sonar consists of M transmitting transducers and N receiving hydrophones. The M-element transmitting transducer forms an M-element transmitting Uniform Linear Array (ULA), and the N receiving hydrophones form an N-element receiving ULA.
Transmitting transducer spacing d t Spaced from the receiving hydrophone by a distance d r Satisfy the requirements of
d t ＝Nd r (3)
At this time, the transmitting ULA and the receiving ULA form an MN-cell large aperture virtual ULA having an array cell spacing d r . The MIMO sonar array is shown in fig. 1.
The MIMO sonar uses M orthogonal code signals having the same frequency band, such as an orthogonal phase code signal, an orthogonal discrete frequency code signal, and the like. The autocorrelation function sidelobes and the cross-correlation function of the coded signals jointly determine range sidelobes in the MIMO sonar imaging output. Coded in quadrature phaseFor the code signal as an example, let the M (M is 1,2, … …, M) th signal be s m (t), the expression of which is:
wherein T represents time, L represents the number of subcodes, T 0 For the length of a single sub-code,
for the initial phase, f, of the l-th sub-code in the m-th transmitted signal 0 To the center frequency of the transmitted signal, j denotes the imaginary part.
Let the region to be imaged in the far field be modelled as P discrete points. Since the MIMO array adopts a dense array mode, the angle from the p (M ═ 1,2, … …, M) th scattering point to all the transmitting and receiving array elements can be considered to be the same. To simplify the analysis, the doppler shift of the echo is negligible, assuming that the relative position between the array and the target is constant. Furthermore, the energy loss due to diffusion and absorption is neglected, and only the influence of the scattering rate of the scattering point on the echo intensity is considered. Accordingly, the received signal on the nth receiving array element can be regarded as the superposition of M mutually independent transmitting signals after different time delays and attenuations, namely
Wherein σ p Is the scattering intensity of the p-th scattering point,
the time delay from the m-th transmitting array element to the p-th scattering point,
the time delay from the p scattering point to the n receiving array element, and n (t) is additive noise.
At the receiving end, the copy of M transmitting signals is used for carrying out matched filtering on the receiving signals on N receiving array elements, and MN matched filtering outputs can be obtained. The (m-1) N + N matched filter outputs are obtained by the matched filter output of the mth transmitting signal copy to the nth receiving array element echo signal, which can be expressed as
y (m-1)N+n (t)＝x n (t)*h m (t) (6)
Wherein represents convolution, h m (t) is the impulse response function of the matched filter corresponding to the mth transmission signal, expressed as
h m (t)＝[s m (T-t)] c (7)
Wherein] c To take conjugation, T ═ LT 0 Is the length of a single transmitted signal.
Since the echo at each receiving array element is a superposition of M independent signals, the output of each matched filter corresponds to 1 autocorrelation function and (M-1) cross-correlation functions. When the transmitted signal is uncorrelated with noise, the matched filter output of the noise n (t) can be omitted, and equation (4) can be written as
Wherein R is m,m (t) is the autocorrelation function of the mth transmitted signal, R m,i As a cross-correlation function between the mth transmit signal and the other transmit signals,
and
has the same meaning, but satisfies i ≠ m.
The transmitting ULA and the receiving ULA in the MIMO sonar form an MN-ary large aperture virtual ULA. The weighting vector can be designed according to the array manifold of the MN element ULA, and the MN matched filtering outputs are processed by beam forming. Let the Q (Q is 1,2, …, Q) th beam output be B q (t) of the formula
Wherein, the first and the second end of the pipe are connected with each other,
the weighting for the (m-1) N + N matched filter outputs,
for corresponding amplitude weighting, theta q Is the scan angle on the qth beam.
And (4) solving absolute values of all beam outputs in the formula (7), and splicing the beam outputs together according to the beam scanning sequence to obtain the original imaging output of the MIMO sonar.
Step 2) mainly relates to a point spread function for deconvolution processing of a Richardson-Lucy algorithm by utilizing orthogonal emission waveforms of MIMO sonar, and the related theory and the specific content are as follows:
according to the formula (6), the matched filter output is composed of an autocorrelation function term and a cross-correlation function term, wherein the cross-correlation function R m,i (t) and an autocorrelation function R m,m And (t) side lobes jointly form range side lobes in the original imaging output of the MIMO sonar. Ideally, the cross-correlation function of the quadrature polyphase coded signal is negligible compared to the peak of the main lobe of the autocorrelation function, while the side lobes of the autocorrelation function are also negligible compared to the peak of the main lobe of the autocorrelation function. Therefore, in an ideal case, the formula (6) can be simplified to
Wherein R is 0 (t) represents an ideal autocorrelation function with flat sidelobe levels. Since the transmitted signal is a quadrature phase encoded signal, R 0 (t) can be represented by
Wherein η represents the point spread function R 0 (t) sidelobe level.
Equation (9) can be further written as a form of expression of convolution
Wherein the content of the first and second substances,
is a dirac function delta (t) at delay
The result of (1). δ (t) takes a value of 0 except where t is 0, and the integral over the entire domain is equal to 1.
From equation (11), ideally, the matched filter output is represented as R 0 (t) and Dirac function
Time domain convolution of (1). In practical situations, since the transmitted orthogonal signals are not perfectly orthogonal and the echo signals have a certain degree of distortion, the autocorrelation function sidelobe and the cross-correlation function in the matched filtering output cannot be ignored. In this case, the autocorrelation function sidelobe and the cross-correlation function can be regarded as the noise term z (t) of the convolution output in expression (11), that is, the autocorrelation function sidelobe and the cross-correlation function
Therefore, as can be seen from equation (12), if we want to suppress the convolution output noise z (t) composed of the autocorrelation function side lobe and the cross correlation function, we need to construct a point spread function and deconvolute it. Because the deconvolution processing can restore the original distribution of the target to the maximum extent and suppress the noise in the convolution output, the noise term z (t) in the formula (12) can be suppressed by the deconvolution processing, thereby achieving the purpose of suppressing the range side lobe. From the analysis of equations (10) to (12), the point spread function PSF (t) can be expressed as
Where η represents the side lobe level of the point spread function psf (t).
The side lobe η of the point spread function in equation (13) needs to be determined before side lobe suppression, and η can be calculated by a search method.
The required substeps are as follows:
substep 1) setting the value of η in equation (13) to a variable interval included from 10 -6 To 10 -0.01 The obtained interval is used for obtaining a plurality of point spread functions PSF (t) under different eta values.
Substep 2) sets the target number in equation (6) to P1 and the scattering coefficient σ to be p Neglecting the time delay term to get the correlation function r (t) ═ 1. R (t) is represented by
And a substep 3) of deconvoluting the absolute value | R (t) | of R (t) in the formula (14) and a plurality of point spread functions at different values of eta as inputs of the Richardson-Lucy algorithm. Where, | | represents solving for an absolute value. The Richardson-Lucy algorithm is an existing algorithm, and deconvolution is an iterative process. In the process of deconvoluting the correlation function R (t) by using Richardson-Lucy algorithm, the (k + 1) th iteration result r (k+1) (t) can be represented by
In the formula (15), as the iteration proceeds, the iteration result r (k) (t) will converge to a unique solution that minimizes the Csszar distance
Wherein the content of the first and second substances,
r (t) representing solving so that the function in parentheses takes the minimum value, and L () representing the Csiszar distance. Assuming that a (x) and b (x) are two variables, the expression for the Cssizar distance is
Substep 4) solving the highest sidelobe level of the output of the Richardson-Lucy algorithm, and finding out the eta value corresponding to the lowest value in all the highest sidelobe levels, wherein the eta value is set as eta value 0 . By η 0 Constructing point spread function PSF (t, eta) for MIMO sonar range sidelobe suppression 0 ) The expression is
Step 3) mainly relates to the deconvolution processing of the MIMO sonar original program imaging output obtained in the step 1) by using the point spread function designed in the step 2) and a Richardson-Lucy algorithm to reduce the distance side lobe, and the related theory and the specific content are as follows:
using the point spread function PSF (t, η) in equation (18) 0 ) And Richardson-Lucy algorithm, deconvoluting the MIMO sonar original imaging output to suppress the equation (12) distance sidelobe term z (t). For distance sidelobe suppression of MIMO sonar, the input of Richardson-Lucy algorithm is point spread function PSF (t, eta) 0 ) Sum beam output absolute value | B q (t) |, output being the target distribution S obtained for the kth iteration (k) (t)。
In the iteration process, the k +1 th iteration result S of the Richardson-Lucy algorithm (k+1) (t) can be represented by
In the formula (19), as the iteration proceeds, S (k) (t) will converge to a unique solution that minimizes the Cssizar distance
Wherein, the first and the second end of the pipe are connected with each other,
represents S (t) for solving so that the function in parentheses takes the minimum value, and L () represents the Cisszar distance.
The point spread function PSF (t, η) in equation (18) is used for the absolute values of all beam outputs 0 ) And performing deconvolution processing on the obtained data by a Richardson-Lucy algorithm to obtain imaging output after sidelobe suppression.
The main steps of the invention are as shown in FIG. 2, calculating the optimal point spread function PSF (t, eta) 0 ) As shown in FIG. 3, using the optimal point spread function PSF (t, η) 0 ) The flow of deconvolution sidelobe reduction processing by the Richardson-Lucy algorithm is shown in FIG. 4.
The embodiment of the invention is given by taking a typical underwater two-dimensional imaging process as an example. The implementation example verifies that the deconvolution-based method can effectively reduce the range sidelobe in the MIMO sonar imaging result from the two-dimensional imaging results of the single scattering point target and the multiple scattering point target respectively.
1) Setting imaging sonar and transmitting signal parameters:
the transmitted signal is assumed to be a sound wave, and the propagation speed of the sound wave under water is 1500 m/s. The MIMO sonar has 3 transmitting transducers and 32 receiving hydrophones, and the spacing of array elements of the receiving hydrophones is lambda/2, wherein lambda corresponds to the wavelength of a 100kHz sound wave signal under water.
The transmission signal of the MIMO sonar is 3 quadrature phase coded signals (see formula (2)), the number of subcodes is 64, the pulse width of the word code is 0.2ms, and the center frequency is 100 kHz.
2) Setting the underwater target position:
imaging simulation is carried out in two times, and underwater target parameters in the two times of simulation are different. In the first simulation, the underwater target is a single scattering point target and is located on coordinates of (0 ° 100 m). In the second simulation, the underwater target is a multi-scattering point target.
3) Designing an optimal point spread function
Setting the value range of eta in the point spread function to be 10 -2 To 10 -0.2 With a variation interval of 10 -0.1 And calculating the distance dimension side lobe level of the original result under different eta values and the distance dimension side lobe level of the deconvolution result. The side lobe levels were obtained by calculating the average of 100 side lobe levels using 100 cycles, and the results are shown in fig. 5 (a). As can be seen from fig. 5(a), when the value of η satisfies log10(η) — 0.7, that is, η ═ 10 -0.7 The lowest range sidelobe level can be obtained by deconvolution using the point spread function. η of fig. 5(b) is 10 -0.7 Fig. 5(c) is a top view of the correlation function r (t) when the loop is cycled 100 times, and fig. 5(d) is a top view of the correlation function r (t) when the loop is cycled 100 times after deconvolution. Comparing fig. 5(c) and fig. 5(d), it can be seen that the proposed deconvolution method can significantly reduce the range-dimension side lobe. From the results of FIG. 5, η is used in the present embodiment 0 ＝10 -0.7 To design an optimal point spread function.
4) And (3) performing imaging treatment:
to reduce the amount of computation, the echo is demodulated using a 92kHz signal, and the demodulated echo is sampled using 40 kHz. The signal-to-noise ratio on the hydrophone was set to 10dB and the noise added was white gaussian noise. The signal-to-noise ratio is defined by the power signal-to-noise ratio, i.e. the ratio of the signal power to the noise, and the noise power is defined by the frequency band level, and the calculation range is 0Hz to 20 kHz. The beam scan range is-45 ° to 45 °, with a scan interval of 0.5 °. In the beam forming process, in order to avoid the influence of beam side lobes in an angle dimension, a Chebyshev window with the side lobe level of-30 dB is used for beam side lobe suppression.
When the target is a single scatter point target, the corresponding imaging result is shown in fig. 6. Fig. 6(a) shows the MIMO sonar original imaging result. As can be seen from fig. 6(a), since the autocorrelation function side lobe and the cross-correlation function are at a high level, the range side lobe of the original imaging result is high. Fig. 6(b) shows the result of the deconvolution method of the present invention. As can be seen from fig. 6(b), when the optimal point spread function and Richardson-Lucy algorithm constructed according to fig. 5 are used, the proposed deconvolution method can effectively suppress range side lobes in MIMO sonar imaging results. Fig. 6(c) shows the distance dimension slice of the original imaging result of the MIMO sonar and the result of the deconvolution method provided by the present invention when the beam scanning angle is 0 °. The MIMO sonar raw imaging result is referred to as MIMO in fig. 6(c), and the deconvolution method result proposed by the present invention is referred to as MIMO-dCv in fig. 6 (c). From the distance dimension slice in fig. 6(c), the sidelobe level of the original imaging result of the MIMO sonar is-13.88 dB, and the sidelobe level of the result of the deconvolution method provided by the present invention is-22.68 dB, which indicates that the method of the present invention can effectively suppress the distance dimension sidelobe in the imaging result of the MIMO sonar.
When the target is a multiple scattering point target, the corresponding imaging results are shown in fig. 7. Fig. 7(a) shows the original imaging result of the MIMO sonar. As can be seen from fig. 7(a), the range sidelobe of the original imaging result is high, which has a bad influence on the imaging quality. Fig. 7(b) shows the result of the deconvolution method of the present invention, and it can be seen from fig. 7(b) that the deconvolution method can effectively suppress the range side lobe in the MIMO sonar imaging result, so that the image becomes clearer. Fig. 7(c) shows the distance dimension projection of the original imaging result of the MIMO sonar and the result of the deconvolution method provided by the present invention. The MIMO sonar raw imaging result is referred to as MIMO in fig. 7(c), and the deconvolution method result proposed by the present invention is referred to as MIMO-dCv in fig. 7 (c). From the distance dimension projection of fig. 7(c), when the target is a multi-scattering point, the side lobe of the original imaging result of the MIMO sonar is about-10 dB, while the side lobe of the result of the deconvolution method provided by the present invention is about-20 dB, which indicates that the method of the present invention can effectively suppress the distance dimension side lobe in the imaging result of the MIMO sonar.
According to an embodiment example, it can be considered that: the deconvolution method provided by the invention can effectively inhibit the autocorrelation function sidelobe and the cross-correlation function of the MIMO sonar, thereby obviously reducing the distance sidelobe in the imaging result.
Claims (2)
1. A deconvolution-based MIMO sonar range sidelobe suppression method is characterized by comprising the following steps:
step S1: obtaining original MIMO sonar imaging output by an MIMO sonar imaging method:
step S2: an optimal point spread function for deconvolution processing is designed from orthogonal waveforms used by the MIMO sonar, including the following sub-steps:
substep S201: preserving the main lobe part of the autocorrelation function in the orthogonal polyphase coding signal used by MIMO sonar, removing the side lobe part of the autocorrelation function, designing a positive number eta as the side lobe level of the designed point spread function, and obtaining the expression of the point spread function PSF (t)
Wherein T represents time, T 0 Represents a single subcode length;
optimizing the value of eta, and setting the value variable interval of eta, which is contained in 10 -6 To 10 -0.01 Obtaining a plurality of point spread functions PSF (t) under different eta values in the formed interval;
substep S202: obtaining a correlation function R (t) according to orthogonal polyphase coding signals used by the MIMO sonar:
wherein R is m,m (t) represents the autocorrelation function of the mth orthogonal polyphase encoded signal, M being 1,2, …, M, R m,i (t) represents the cross-correlation function between the mth orthogonal multi-phase coded signal and other orthogonal multi-phase coded signals, wherein M represents the number of orthogonal multi-phase coded signals used by the MIMO sonar and satisfies that M is more than or equal to 2;
substep S203: obtaining | R (t) | by solving the absolute value of the correlation function R (t), wherein | | represents the absolute value; will be provided withTaking | R (t) | and a point spread function PSF (t) as the input of the Richardson-Lucy algorithm, and performing deconvolution processing; the Richardson-Lucy algorithm is an iterative algorithm, and the k +1 th iteration result r (k+1) (t) is represented by
Wherein k represents the number of iterations;
as the iteration progresses, the iteration result r (k) (t) will converge to a unique solution that minimizes the Csiszar distance, corresponding to the expression:
wherein, the first and the second end of the pipe are connected with each other,
r (t) representing solving such that the function in parentheses takes the minimum value, L () representing the Csszar distance, assuming that a (x) and b (x) are two variables, the expression for the Csszar distance is
Substep S204: repeating the substep S203 to obtain the output of the Richardson-Lucy algorithm under all eta values; solving the highest sidelobe level of the outputs of the Richardson-Lucy algorithms, and finding out the eta value corresponding to the lowest value of all the highest sidelobe levels to be eta value 0 (ii) a By η 0 Constructing point spread function PSF (t, eta) for MIMO sonar distance sidelobe suppression 0 ) The expression is
Step S3: deconvoluting the original imaging output of the MIMO sonar obtained in the step S1 by using the optimal point spread function and the Richardson-Lucy algorithm designed in the step S2 to reduce the range side lobe, wherein the method comprises the following sub-steps:
substep S301: the input of the Richardson-Lucy algorithm is an optimal point spread function PSF (t, eta) 0 ) And the q-th beam output B q Absolute value | B of (t) q (t) |, Q ═ 1,2, …, Q, with the output being the target distribution S obtained for the kth iteration (k) (t)；
Substep S302: in the iterative process of reducing distance sidelobe, the k +1 iteration result S of the Richardson-Lucy algorithm (k+1) (t) is represented by
As the iteration progresses, S (k) (t) will converge to a unique solution that minimizes the Cssizar distance
Wherein the content of the first and second substances,
(t) represents solving s (t) such that the function in parentheses takes the minimum value;
substep S303: repeating the substep S302 using the point spread function PSF (t, η) for the absolute values of all beam outputs 0 ) And performing deconvolution processing by a Richardson-Lucy algorithm to obtain imaging output after sidelobe suppression.
2. The deconvolution-based MIMO sonar distance sidelobe suppression method of claim 1, wherein said obtaining MIMO sonar imaging output in step S1 includes the following sub-steps:
substep S101: defining MIMO sonar including M transmitting transducers and N receiving hydrophones, M-element transmittingThe transmitting transducer forms an M-element transmitting uniform linear array, and the N receiving hydrophones form an N-element receiving uniform linear array; transmitting transducer spacing d t Spaced from the receiving hydrophone by a distance d r Satisfy the requirements of
d t ＝Nd r ；
Substep S102: the MIMO sonar uses M orthogonal coding signals with the same frequency band, M is more than or equal to 2, and the mth signal is s m (t), M ═ 1,2, … …, M, expressed as:
wherein T represents time, L represents the number of subcodes, T 0 Is the length of a single sub-code,
for the initial phase, f, of the l-th sub-code in the m-th transmitted signal 0 J represents the imaginary part for the center frequency of the transmitted signal;
substep S103: let the region to be imaged in the far field be modeled as P discrete points, and the received signal on the nth receiving array element be:
wherein σ p Is the scattering intensity of the p-th scattering point,
the time delay from the m-th transmitting array element to the p-th scattering point,
is the time delay from the p scattering point to the n receiving array element, and n (t) is additive noise;
substep S104: neglecting the matched filtering output of the noise N (t), performing matched filtering on the receiving signals on the N receiving array elements by using copies of M transmitting signals at a receiving end to obtain MN matched filtering outputs, wherein the (M-1) N + N matched filtering outputs are obtained by outputting matched filtering of the mth transmitting signal copy on the echo signal on the nth receiving array element and are expressed as
R m,m (t) is the autocorrelation function of the mth orthogonal polyphase encoded signal, R m,i (t) is a cross-correlation function between the mth orthogonal polyphase coded signal and the other orthogonal polyphase coded signals,
representing the time delay from the ith transmitting array element to the pth scattering point, and meeting the condition that i is not equal to m;
substep S105: performing beam forming processing on the MN matched filter outputs; let the qth beam output be B q (t)
Wherein, the first and the second end of the pipe are connected with each other,
for the (m-1) N + N matched filter outputs,
for corresponding amplitude weighting, theta q Is the scan angle on the qth beam;
will be provided with
Obtaining | B by solving absolute value of all beam outputs q (t) |, and spliced together according to the beam scanning sequence to obtainTo the original imaging output of the MIMO sonar.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010083253.5A CN111257891B (en) | 2020-02-08 | 2020-02-08 | Deconvolution-based MIMO sonar distance sidelobe suppression method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010083253.5A CN111257891B (en) | 2020-02-08 | 2020-02-08 | Deconvolution-based MIMO sonar distance sidelobe suppression method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111257891A CN111257891A (en) | 2020-06-09 |
| CN111257891B true CN111257891B (en) | 2022-07-26 |
Family
ID=70952699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010083253.5A Active CN111257891B (en) | 2020-02-08 | 2020-02-08 | Deconvolution-based MIMO sonar distance sidelobe suppression method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111257891B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111812612A (en) * | 2020-06-18 | 2020-10-23 | 西安电子科技大学 | Partial correlation waveform design method of MIMO radar based on subarray orthogonal LFM signal |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008105909A2 (en) * | 2006-07-28 | 2008-09-04 | University Of Florida | Optimal beam pattern synthesis via matrix weighting |
| CN101406041A (en) * | 2006-05-08 | 2009-04-08 | 三菱电机株式会社 | Method for reducing blur in an image of a scene and method for deblurring an image of a scene |
| CN108761394A (en) * | 2018-04-17 | 2018-11-06 | 哈尔滨工程大学 | A kind of high-resolution low sidelobe based on space-time processing deconvolutes Power estimation method |
| CN109375227A (en) * | 2018-11-30 | 2019-02-22 | 中国科学院声学研究所 | A kind of deconvolution Wave beam forming three-dimensional acoustic imaging method |
| CN109407098A (en) * | 2018-08-30 | 2019-03-01 | 中国船舶重工集团公司第七〇五研究所 | A kind of implementation method for the parallel processing algorithm handling orthogonal pulses string signal |
| CN110412587A (en) * | 2019-07-19 | 2019-11-05 | 中国科学院声学研究所 | A kind of lower view synthetic aperture three-D imaging method and system based on deconvolution |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7106250B2 (en) * | 2003-09-03 | 2006-09-12 | The United States Of America As Represented By The Secretary Of The Navy | Robust predictive deconvolution system and method |
| WO2017083597A1 (en) * | 2015-11-13 | 2017-05-18 | Flir Systems, Inc. | Detection and ranging systems and methods |
-
2020
- 2020-02-08 CN CN202010083253.5A patent/CN111257891B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101406041A (en) * | 2006-05-08 | 2009-04-08 | 三菱电机株式会社 | Method for reducing blur in an image of a scene and method for deblurring an image of a scene |
| WO2008105909A2 (en) * | 2006-07-28 | 2008-09-04 | University Of Florida | Optimal beam pattern synthesis via matrix weighting |
| CN108761394A (en) * | 2018-04-17 | 2018-11-06 | 哈尔滨工程大学 | A kind of high-resolution low sidelobe based on space-time processing deconvolutes Power estimation method |
| CN109407098A (en) * | 2018-08-30 | 2019-03-01 | 中国船舶重工集团公司第七〇五研究所 | A kind of implementation method for the parallel processing algorithm handling orthogonal pulses string signal |
| CN109375227A (en) * | 2018-11-30 | 2019-02-22 | 中国科学院声学研究所 | A kind of deconvolution Wave beam forming three-dimensional acoustic imaging method |
| CN110412587A (en) * | 2019-07-19 | 2019-11-05 | 中国科学院声学研究所 | A kind of lower view synthetic aperture three-D imaging method and system based on deconvolution |
Non-Patent Citations (5)
| Title |
|---|
| Lei XIE et al..A Deconvolved method for suppressing the sidelobes in the DOA Estimation.《2018 OCEANS - MTS/IEEE Kobe Techno-Oceans (OTO)》.2018, * |
| Xionghou Liu et al..Low Sidelobe Range Profile Synthesis for Sonar Imaging Using Stepped-Frequency Pulses.《IEEE Geoscience and Remote Sensing Letters 》.2016,第14卷(第2期), * |
| Xueli Sheng et al..An Interference Suppression Technique Based on Deconvolution and Null Constraint.《OCEANS 2018 MTS/IEEE Charleston》.2019, * |
| 王朋等.二维解卷积波束形成水下高分辨三维声成像.《声学学报》.2019,第44卷(第4期), * |
| 许小剑等.MIMO雷达三维干涉诊断成像方法.《雷达学报》.2018,第7卷(第6期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111257891A (en) | 2020-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107104720B (en) | Mutual-prime array self-adaptive beam forming method based on covariance matrix virtual domain discretization reconstruction | |
| WO2011008146A1 (en) | Method and wideband antenna system to minimise the influence of interference sources | |
| CN109143190B (en) | Broadband steady self-adaptive beam forming method for null broadening | |
| CN109490850A (en) | Wideband array Adaptive beamformer method under major lobe suppression | |
| CN109597041B (en) | Segmented linear frequency modulation waveform design method based on coherent FDA | |
| CN113325385B (en) | Anti-interference method for phased array-MIMO radar mode transmit-receive beam forming | |
| Stergiopoulos | Implementation of adaptive and synthetic-aperture processing schemes in integrated active-passive sonar systems | |
| Kozick et al. | Synthetic aperture pulse-echo imaging with rectangular boundary arrays (acoustic imaging) | |
| Chi | Underwater Real-Time 3D Acoustical Imaging: Theory, Algorithm and System Design | |
| CN111693971A (en) | Wide beam interference suppression method for weak target detection | |
| CN111257891B (en) | Deconvolution-based MIMO sonar distance sidelobe suppression method | |
| Liu et al. | High-resolution and low-sidelobe forward-look sonar imaging using deconvolution | |
| CN110196421B (en) | Dense MIMO sonar self-adaptive beam forming detection method | |
| Liu et al. | Using deconvolution to suppress range sidelobes for MIMO sonar imaging | |
| CN111239747B (en) | Sonar high-resolution low-sidelobe two-dimensional imaging method based on deconvolution | |
| Hum et al. | UWB beamforming using 2-D beam digital filters | |
| JP2006284241A (en) | Filter apparatus | |
| CN108761433B (en) | High-resolution imaging method using MIMO sonar difference array processing | |
| CN109669172B (en) | Weak target direction estimation method based on strong interference suppression in main lobe | |
| CN108919239B (en) | Ultrasonic direct wave cancellation method based on double-combination adaptive filtering | |
| KR100922180B1 (en) | Method of Optimum design of periodic sparse arrays for ultrasound image system | |
| CN116148777A (en) | Array radar main lobe interference suppression method based on array virtual expansion | |
| CN113406562B (en) | TOA and DOA combined estimation dimension reduction method in Beidou and ultra-wideband system | |
| Liu et al. | Compensating for intensity loss in a large-aperture MIMO sonar imaging system | |
| CN113704685A (en) | Deep sea blind deconvolution method based on vertical line array |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |