CN1800874A - Signal processing method by using sonar to measure target - Google Patents

Abstract

The invention discloses a signal processing method by measuring target with sonar, which comprises: constructing a time-space correlation function matrix RTT with a signal matrix X formed by target echo signal received by the sonar, a left real transformation matrix BM, and the operation to let the RTT as real matrix; measuring the target with signal subspace method according to said RTT. This invention reduces calculation amount and has super performance to the prior art.

Description

A kind of signal processing method with sonar to measure target

Technical field

The present invention relates to the sonar field, more particularly, the present invention relates to a kind of signal processing method with sonar to measure target.

Background technology

In the last few years, the high resolving power beam-forming technology development in the modern battle array signal processing technology was very fast.The high resolving power beam-forming technology utilizes the time-space correlation function of transducer array to constitute matrix, and matrix extracts the information of sound wave thus, comprises sound wave incident angle and amplitude etc.Except echo signal, also has noise signal in the signal that transducer array receives.Therefore transducer array time-space correlation function matrix can resolve into signal subspace and noise subspace in functional space, and two sub spaces are vertical mutually.Correspondingly, to the general two big classes of dividing of signal processing method of transducer array time-space correlation function matrix, a class is the spectrum based method, it comprises the noise subspace method, claim Zero Space Method and Its again, when small sample, low signal-to-noise ratio and high signal coherency, the performance of these class methods obviously descends.Another kind of is parametric method, and it comprises the signal subspace method.The performance of parametric method obviously is better than composing based method.

In the prior art, the noise subspace method in the high resolving power beam-forming technology has had application in sonar.For example, United States Patent (USP) 6130641 " Imaging methods and apparatus usingmodel-based array signal processing " people such as P.Kraeutner, and people's such as people's such as P.Kreautner article " Principlecomponents array processing for swath acoustic mapping; proceedings of IEEEOceans ' 97 Conference; October 1997 " and H.Kreautner article " Beyondinterferometry; resolving multiple angles-of-arrival in swath bathymetric imaging; proceedings of the IEEE Ocean ' 99 Conference; September, 1999 " in; adopt the noise subspace method that transducer array time-space correlation function matrix is handled, obtained the resolution higher than conventional beam-forming technology.

In the above-mentioned document of quoting, its signal processing method has following problem:

1) the time-space correlation function matrix only is made of signal matrix usually, and signal matrix is made up of the received signal of sonar.If X represents signal matrix with matrix, then time-space correlation function is generally X ^HThe form of X, wherein subscript H represents conjugate transpose, because signal matrix X complex matrix, so its temporal and spatial correlations Jacobian matrix also is a complex matrix.Such time-space correlation function does not carry out decorrelation to signal to be handled, and when there be high being concerned with in signal, the sonar performance descended.And, be to carry out to the processing of time-space correlation function matrix in complex space, its operand is very big.

2) the noise subspace method that adopts the high resolving power wave beam to form in the signal Processing is handled the time-space correlation function matrix, and noise subspace method performance when small sample, low signal-to-noise ratio and high signal coherency obviously descends.

3) existing signal processing method obtains reasonable effect when measuring marine target (target in the water in other words), but effect is relatively poor when measuring sub-sea floor targets (water-bed in other words target).In the article of for example aforementioned P.Kreautner that quotes and H.Kreautner, in the pond man-made target is measured, target is that the copper pipe by quadrature constitutes, and this is a good acoustic target, sonar can correctly detect man-made target, but poor to the measurement result of pool wall.

4) the resolution wave beam forms usually to run in the signal Processing and separates system of linear equations, and will weaken The noise, and least square method is the available general method of performance, and it has considerable restriction to the type and the interference of noise form of noise.

5) orientation of the root estimating target of the polynomial expression group corresponding in the usefulness kernel with the time-space correlation function matrix, this method has shortcoming, and at first, it has all estimated the signal that incides on the transducer array, do not select desired direct wave signal, do not delete many ways undesired signals yet; Secondly, have only the precision ability of just determining the target azimuth of trying to achieve after the number of targets high, a root estimating target number, and while estimating target orientation, the target Bearing Estimation precision that this is difficult to obtain under some situations with the polynomial expression group.

Summary of the invention

One object of the present invention is to provide a kind of signal processing method with sonar to measure target, has constructed a new time-space correlation function matrix in the method; Another object of the present invention is to provide a kind of signal processing method with sonar to measure target, this method is carried out target measurement with the signal subspace method.

In order to realize the foregoing invention purpose, the invention provides a kind of signal processing method with sonar to measure target, described target is marine target or sub-sea floor targets, described signal processing method comprises structure time-space correlation function matrix step and according to described time-space correlation function matrix measurement target step, wherein:

In structure time-space correlation function matrix step, described time-space correlation function matrix Comprise signal matrix X and and the left consolidation that multiplies each other of signal matrix change matrix B _M, also comprise making described time-space correlation function matrix The real arithmetic of getting for real matrix; Described signal matrix is the complex matrix that is made of the target echo signal that sonar receives;

In the measurement target step, adopt the signal subspace method to measure described target.

Described left consolidation changes matrix B _MBe the matrix of a M * M dimension, its form is: $B_M=\left[\begin{array}{cc}{I}_{M/2}& j{I}_{M/2}\\ P_{M/2}& -j{P}_{M/2}\end{array}\right],$ Wherein M is a sonar transducer battle array primitive number, I _M/2Be M/2 rank unit matrix, R _M/2Be M/2 rank symmetry permutation matrix.

Described time-space correlation function matrix ${\hat{R}}_{\mathrm{TT}}=\frac{1}{\mathrm{MN}}\mathrm{Re}\left[{2B}_M^H{\mathrm{XX}}^H{B}_M\right],$ Wherein M is a sonar transducer battle array primitive number, and N is the number of samples of sonar in an equivalent pulsewidth, and Re represents to get practical operation and does, and subscript H represents the transpose conjugate computing.

In the measurement target step, comprising: calculate the time-space correlation function matrix

Eigenwert; According to the time-space correlation function matrix Eigenwert estimating target number; According to estimated target number signal calculated subspace

In signal subspace, obtain the direction and/or the signal waveform of each target.Wherein, in signal subspace, adopt total least square method to obtain the direction of each target.In signal subspace, adopt the signal restoring algorithm to obtain the signal waveform of each target.When described target is sub-sea floor targets, also comprise step according to the direction calculating sea floor height of target.When described target is sub-sea floor targets, also comprises according to the submarine topography Changing Pattern and judge whether really step of described sea floor height; If judge that described sea floor height is untrue, also comprise according to the time-space correlation function matrix

The eigenwert step of estimating target number once more.

Described signal matrix is to be made of the signal that the echoed signal that sonar receives surpasses averaged magnitude.

Advantage of the present invention:

1) in the present invention, the time-space correlation function matrix changes matrix by signal matrix and left real-turn and unites and constitute, and make that by getting practical operation the time-space correlation function matrix is a real matrix, the related function matrix of Gou Chenging is understood relevant treatment to signal like this, even the performance that received signal still can obtain when having high being concerned with.And, be to carry out to the processing of time-space correlation function matrix in the real space, obviously reduced operand.

2) in the present invention, the signal subspace method that adopts the high resolving power wave beam to form in the signal Processing is handled the time-space correlation function matrix, and signal subspace method performance when small sample, low signal-to-noise ratio and high signal coherency obviously is better than the noise subspace method.

3) than existing signal processing method, the present invention has more excellent effect when measuring sub-sea floor targets (water-bed in other words target).

Description of drawings

Fig. 1 is the composition synoptic diagram that adopts the high-resolution three-dimension side scan sonar system of signal processing method of the present invention;

Fig. 2 is the concrete composition structural drawing of high-resolution three-dimension side scan sonar system shown in Figure 1;

Fig. 3 is the process flow diagram of an embodiment of signal processing method of the present invention;

Fig. 4 is that the high-resolution three-dimension side scan sonar system adopts the test findings of signal processing method of the present invention in the pond;

Fig. 5 is contained in the high-resolution three-dimension side scan sonar system to adopt the test data sheet of signal processing method of the present invention in the deep water lake on the underwater robot;

Fig. 6 is that the high-resolution three-dimension side is swept the water-bed mima type microrelief landforms that the acoustic image system adopts signal processing method of the present invention to obtain.

Embodiment

Below in conjunction with the drawings and specific embodiments the present invention is described in further detail.

Fig. 1 shows the composition synoptic diagram of the high-resolution three-dimension side scan sonar system that adopts signal processing method of the present invention.This high-resolution three-dimension side scan sonar system comprises transducer array 200 and electronics extension set 300, and they are arranged on the carrier 100 that is positioned under the sea 104.The terminal computer 400 of high-resolution three-dimension side scan sonar system is arranged on the lash ship (not shown) on the sea 104.Be connected by cable between terminal computer 400 and the electronics extension set 300.The high-resolution three-dimension side scan sonar system passes through transducer array 200 to seabed launching beam 101, and the reflection echo signal of reception such as targets such as marine target (not shown) or seabed 102 and/or sub-sea floor targets 103, after handling, electronics extension set 300 obtains metrical informations such as each target direction angle.As shown in Figure 1, the fan-shaped beam 101 that transducer array 200 sends wave beam in vertical plane is very wide, and wave beam is very narrow in surface level.

The high-resolution three-dimension side scan sonar system is formed as shown in Figure 2 more specifically.Transducer array 200 comprises left receiving transducer 201, left transmitting transducer 203, right receiving transducer 202 and right transmitting transducer 204.Electronics extension set 300 comprises controller 307, main control computer 308, attitude sensor 309, temperature sensor 310 and the hard disk 311 of the process that left receiver 301, left transmitter 303, right receiver 302, right transmitter 304, multi-channel a/d converter 305, high speed digital signal processor 306, control system transmit and receive.Electronics extension set 300 links by Ethernet 312 and terminal computer 400.

As shown in Figure 2, left transmitter 203 in the electronics extension set 300 and right transmitter 204 respectively with transducer array 200 in left transmitting transducer 203 be connected with right transmitting transducer 204, the left receiver 301 in the electronics extension set 300 and right receiver 302 respectively with transducer array 200 in left receiving transducer 201 be connected with right receiving transducer 202.Left side receiver 301 is connected with hyperchannel A/D transducer 305 with right receiver 302, multi-channel a/d converter 305 is connected with high speed digital signal processor 306, high speed digital signal processor 306 is connected with main control computer 308, main control computer 308 is connected with hard disk 311 with controller 307, and is connected with terminal computer 400 through Ethernet 312.Controller 307 is connected with left receiver 310, right receiver 302, left transmitter 303, right transmitter 304, attitude sensor 309 and temperature sensor 310.

The special measurement program of high-resolution three-dimension side scan sonar system illustrated in figures 1 and 2 is loaded in the storer of main control computer 308, and presses the step in the process flow diagram shown in Fig. 3 and carry out.

The 401st, the beginning step is sent instruction by terminal computer 400, is transferred to the main control computer 308 in the electronics extension set 300, starts the program in the computing machine 308, makes sonar in running order.

In step 402 and 403, the software and hardware of sonar system is carried out initialization by initialization module.

In step 404, main control computer 308 generates and transmits, and normally obtains the digitized waveform that transmits with computing machine.Transmitting here preferably has narrow equivalent pulsewidth, makes the equivalent pulsewidth that transmits less than 0.1 millisecond in actual applications usually.The equivalent pulsewidth that transmits is narrow more, and the area of space of signal representative that arrives sonar a moment is narrow more, can reduce the target echo number that different directions arrives transducer array simultaneously like this).Obtain narrow equivalent pulsewidth in order to make to transmit, transmitting preferably has narrow related function, and such signal comprises narrow simple pulse and linear FM signal (Chirp signal) pulse.

In step 405, main control computer 308 via controllers 307 drive transmitter 303 and 304, drive transmitting transducer 203 and 204 again, and the emission sound pulse arrives in the fluid media (medium) (for example seawater) towards the seabed.

In step 406, receiving transducer 201 and the 202 backscattered echoed signals that receive from fluid media (medium), marine target and sub-sea floor targets, and present to receiver 301 and 302.

In step 407, carry out demodulation filtering by receiver 301 and 302 pairs of echoed signals, send into multi-channel a/d converter 305 then.

In step 408, multi-channel a/d converter 305 becomes digital signal with echoed signal from analog signal conversion, delivers in the high speed digital signal processor 306, therein execution in step 409～step 416 one by one again.

In step 409, calculate the mean intensity of received signal, and pick out the received signal of signal intensity greater than mean intensity.

In step 410, the time-space correlation function matrix of a reality of structure

With the received signal structure signal matrix X that picks out, signal matrix X is the complex matrix of being tieed up by a M * N that the backscattering echoed signal constitutes, and M is a transducer array primitive number, and generally getting M is even number, and N is the number of samples in an equivalent pulsewidth.In the present invention, except signal matrix X, also provide a left consolidation to change matrix B in the time-space correlation function matrix _M, in one embodiment, it is B that this left side consolidation changes matrix _MBe the matrix of a M * M dimension, its form is: $B_M=\left[\begin{array}{cc}{I}_{M/2}& j{I}_{M/2}\\ P_{M/2}& -j{P}_{M/2}\end{array}\right],$ As seen B _MCan be decomposed into 2 * 2 partitioned matrix, each submatrix is that M/2 * M/2 ties up matrix, its first behavior submatrix I _M/2And jI _M/2, the second behavior jP _M/2With-jP _M/2, I wherein _M/2Be M/2 rank unit matrix, P _M/2Be M/2 rank symmetry permutation matrix.

In one embodiment, should change matrix B by left side consolidation _MMultiply each other with signal matrix X after doing the conjugate transpose computing, to form a new signal matrix $X^{\′}=B_M^{H}X,$ Constitute the time-space correlation function matrix by this new signal matrix X ' with conventional method again, promptly be configured to X ' X ' ^HForm, wherein subscript H represents the conjugate transpose computing.A preferred exemplary according to the time-space correlation function of the reality of the present invention structure is ${\hat{R}}_{\mathrm{TT}}=\frac{1}{\mathrm{MN}}\mathrm{Re}\left[{2B}_M^H{\mathrm{XX}}^H{B}_M\right],$ Re represents to get practical operation and does.

In the present invention, change matrix B by left consolidation _MWith get practical operation and do, the received signal that is comprised in the pair correlation function matrix is carried out decorrelation and is handled, and makes the performance that the present invention still can obtain when high signal coherency.Simultaneously, because time-space correlation function matrix of the present invention

Be real matrix, in follow-up processing, be real arithmetic, obviously reduced operand.

In step 411～step 414, according to structure real-time empty related function matrix Adopt the signal subspace method to measure each target.

In step 411, calculate real-time empty related function matrix Eigenwert.The compute matrix bundle

Generalized character decompose ∑ wherein _MStructure matrix for noise correlation matrix.The computing formula that generalized character decomposes is: ${\hat{R}}_{\mathrm{TT}}\stackrel{\‾}{E}={\mathrm{\Σ}}_M\stackrel{\‾}{E}\widehat{\mathrm{\Λ}},$ Generalized character value matrix in the formula.∑ _M、 $\widehat{\mathrm{\Λ}}=\mathrm{diag}\{{\widehat{\mathrm{\λ}}}_I\·\·\·\·\·\·{\widehat{\mathrm{\λ}}}_M\},$ ${\widehat{\mathrm{\λ}}}_I\≥\·\·\·\·\·\·\≥{\widehat{\mathrm{\λ}}}_M,$ The generalized character vector matrix $\stackrel{\‾}{E}=\left[{\hat{e}}_1\right|\·\·\·\·\·\·\left|{\hat{e}}_M\right].{\mathrm{\Σ}}_M,$ With

All be the matrix of M * M dimension, M is a transducer array primitive number.

In step 412, according to real-time empty related function matrix

Eigenwert estimate the target number.At the generalized character value matrix $.\widehat{\mathrm{\Λ}}=\mathrm{diag}\{{\widehat{\mathrm{\λ}}}_1\·\·\·\·\·\·{\widehat{\mathrm{\λ}}}_M\}$ In, owing to represent the eigenwert of the eigenwert of noise subspace much smaller than the representation signal subspace, therefore can be by the generalized character matrix $\widehat{\mathrm{\Λ}}=\mathrm{diag}\{{\mathrm{\λ}}_1\·\·\·\·\·\·{\mathrm{\λ}}_M\}$ In ${\widehat{\mathrm{\λ}}}_j(1\≤j\≤M)$ Be worth big

Number number of targets n according to a preliminary estimate.Owing to difference in size between the eigenwert of the eigenwert of representing noise subspace and representation signal subspace is very big, so this is that those skilled in the art is easy to judge.

In step 413, according to estimated target number n signal calculated subspace

Signal subspace ${\hat{S}}_X=R\left\{{\hat{E}}_S\right\},$ R{} is an image space, ${\hat{E}}_S={\mathrm{\&Sigma;}}_M\left[{\mathrm{<figure-callout\; id="1"\; label="e"\; filenames="A20041010403300141.png"\; state="\{\{state\}\}">e</figure-callout>}}_1\right|\&CenterDot;\&CenterDot;\&CenterDot;\&CenterDot;\&CenterDot;\&CenterDot;\left|e_n\right]$ Be by the base of the definite signal subspace of target number n, that is to say signal subspace Be by The image space of opening.

In step 414, in signal subspace, adopt total least square method to obtain the direction and the signal waveform of each target.Particularly, at first calculate ${\hat{E}}_X=C_1{\hat{E}}_S,{\hat{E}}_Y=C_2{\hat{E}}_S,$ Wherein ${\mathrm{<figure-callout\; id="1"\; label="C"\; filenames="A20041010403300141.png"\; state="\{\{state\}\}">C</figure-callout>}}_1=2\mathrm{Re}\left\{B_{M-1}^H{G}_2{B}_M\right\},$ ${\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="A20041010403300141.png"\; state="\{\{state\}\}">C</figure-callout>}}_2=2/m\left\{B_{M-1}^H{G}_2{B}_M\right\},$ G ₂Be (M-1) * Metzler matrix, first classifies zero vector as, and other is unit matrix, wherein a B _M-1Change matrix for the left real-turn of (M-1) * (M-1), be defined as $\left[\begin{array}{ccc}{I}_{(M-1)/2}& 0& {\mathrm{jI}}_{(M-1)/2}\\ 0^{\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="A20041010403300141.png"\; state="\{\{state\}\}">T</figure-callout>}}& \sqrt{2}& 0^T\\ P_{(M-1)/2}& 0& {-\mathrm{jP}}_{(M-1)/2}\end{array}\right],$ Wherein 0 is (M-1)/1 * 1 dimension null matrix.With overall square law solving equation ${\hat{E}}_X\mathrm{\&Gamma;}={\hat{E}}_Y$ After obtain Γ, the eigenwert of compute matrix Γ and left eigenvector matrix then, eigenwert is designated as λ ₁λ _n, the left eigenvector matrix is designated as W.Obtain the deflection of each target at last ${\widehat{\mathrm{\&theta;}}}_k=\arcsin \{-c{\widehat{\mathrm{\&phi;}}}_k/{\mathrm{\&omega;}}_0\mathrm{\&Delta;}\},$ Wherein ${\widehat{\mathrm{\&phi;}}}_k=2\arctan \left({\mathrm{\&lambda;}}_k\right),$ K=1 ... n, c, ω ₀With Δ be respectively the velocity of sound, signal center frequency and transducer array primitive spacing.Adopt the signal restoring algorithm, obtain signal waveform $\hat{S}={\left(W^H{\hat{E}}_s^H{\hat{E}}_{s}W\right)}^{-1}{W}^H{\hat{E}}_s^H{B}_M^{H}X.$

Can obtain the direction and the signal waveform of each target by abovementioned steps.But well-known, in the sonar field, that measures sub-sea floor targets will be difficult to measure marine target, can measure the sonar of sub-sea floor targets and also can measure marine target usually, and the sonar that can measure marine target not necessarily can be measured sub-sea floor targets.As concrete an application of measurement sub-sea floor targets of the present invention, will continue hereinafter to describe by measuring the step that sub-sea floor targets obtains the seabed acoustic image.

After step 414, can obtain the direction and the signal waveform of the target of each sub-sea floor targets.In step 415, according to the sub-sea floor targets arrival direction Try to achieve the height H everywhere and the horizontal range in seabed, the just height of each sub-sea floor targets and horizontal range with sound wave path difference back and forth.Path difference is obtained by the difference in the moment that transmits and the moment that receives certain direction echo.In measuring process, carrier 100 constantly advances in the water, transducer array 200 constantly transmits, seabed involuting wave is received by transducer array 200 according to the priority of time, after handling, electronics extension set 300 can obtain the height of each position, seabed (being sub-sea floor targets) like this, thereby form the acoustic image in seabed, and give terminal computer 400 result transmission.

In step 416, calculate the Changing Pattern of the height H in seabed with horizontal range, with the variation and comparing of generally acknowledging of the height that obtains such as submarine topography Changing Patterns such as slope (submarine topography is no more than 30 degree) and periods of change with distance, if meet, then this assert that this H value is real, changes the acoustic image that step 417 obtains the seabed over to; If do not meet, then change in the step 412, new number of targets n is set, execution in step 413～step 416 obtains the correct degree of depth until last most sub-sea floor targets sample standard deviations once more.

In step 417, the release signal of the target that obtains according to step 414, the intensity of the position echoed signal that recovery is obtained is swept the result as the side of this position, and it is presented on the X-Y scheme, make the actual displayed position consistent, thereby obtain the seabed acoustic image with the actual position of target.

Fig. 4 is that the high-resolution three-dimension side scan sonar system adopts signal processing method of the present invention test findings in the pond, and measurement result and actual pool wall quite meet.

Fig. 5 is contained in the high-resolution three-dimension side scan sonar system to adopt the test data sheet of signal processing method of the present invention in the deep water lake on the underwater robot.Navigation route is a cruciform, the total accuracy of sounding of checking acoustic image system.Measurement result is illustrated in the table 1, can see that by table 1 total accuracy of sounding is 0.5～0.7%, is better than the standard 1% of international navigational route office.

Fig. 6 is that high-resolution three-dimension side of the present invention is swept the water-bed mima type microrelief landforms that the acoustic image system obtains.

Claims (9)

1, a kind of signal processing method with sonar to measure target, described target is marine target or sub-sea floor targets, described signal processing method comprises structure time-space correlation function matrix step and according to described time-space correlation function matrix measurement target step, it is characterized in that:

In structure time-space correlation function matrix step, described time-space correlation function matrix

Comprise signal matrix X and and the left consolidation that multiplies each other of signal matrix change matrix B _M, also comprise making described time-space correlation function matrix The real arithmetic of getting for real matrix; Described signal matrix is the complex matrix that is made of the target echo signal that sonar receives;

In the measurement target step, adopt the signal subspace method to measure described target.

2, the signal processing method with sonar to measure target according to claim 1 is characterized in that described left consolidation changes matrix B _MBe the matrix of a M * M dimension, its form is: $B_M=\left[\begin{array}{cc}{I}_{M/2}& {\mathrm{jI}}_{M/2}\\ P_{M/2}& -j{P}_{M/2}\end{array}\right],$ Wherein M is a sonar transducer battle array primitive number, I _M/2Be M/2 rank unit matrix, P _M/2Be M/2 rank symmetry permutation matrix.

3, the signal processing method with sonar to measure target according to claim 1 and 2 is characterized in that described time-space correlation function matrix ${\hat{R}}_{\mathrm{TT}}=\frac{1}{\mathrm{MN}}\mathrm{Re}\left[2B_M^H{\mathrm{XX}}^H{B}_M\right],$ Wherein M is a sonar transducer battle array primitive number, and N is the number of samples of sonar in an equivalent pulsewidth, and Re represents to get practical operation and does, and subscript H represents the transpose conjugate computing.

4, the signal processing method with sonar to measure target according to claim 1 is characterized in that, in the measurement target step, comprising:

Calculate the time-space correlation function matrix

Eigenwert;

According to the time-space correlation function matrix

Eigenwert estimating target number;

According to estimated target number signal calculated subspace

In signal subspace, obtain the direction and/or the signal waveform of each target.

5, the signal processing method with sonar to measure target according to claim 4 is characterized in that, adopts total least square method to obtain the direction of each target in signal subspace.

6, the signal processing method with sonar to measure target according to claim 4 is characterized in that, adopts the signal restoring algorithm to obtain the signal waveform of each target in signal subspace.

7, the signal processing method with sonar to measure target according to claim 4 is characterized in that, when described target is sub-sea floor targets, also comprises the step according to the direction calculating sea floor height of target.

8, whether really the signal processing method with sonar to measure target according to claim 7 is characterized in that, when described target is sub-sea floor targets, also comprises according to the submarine topography Changing Pattern and judge described sea floor height step; If judge that described sea floor height is untrue, also comprise according to the time-space correlation function matrix

The eigenwert step of estimating target number once more.

9, the signal processing method with sonar to measure target according to claim 1 is characterized in that, described signal matrix is to be made of the signal that the echoed signal that sonar receives surpasses averaged magnitude.

Images