CN102608609A - MIMO (multiple input multiple output) based high-resolution underwater target detection device and method - Google Patents

Abstract

The invention discloses an MIMO (multiple input multiple output) based high-resolution underwater target detection device and method. According to the invention, a potential target is actively detected through a plurality of mutually-orthogonal detection waveforms transmitted by a transmitting transducer array in a hydroacoustic environment, and a hydrophone-array received target echo is received. Through MIMO processing, a method for carrying out detection and estimation on an MIMO frame is established, and the extension of waveform diversity gains and virtual array apertures is obtained. By using the MIMO based underwater target detection method disclosed by the invention, in the process of carrying out hydroacoustic target detection, compared with conventional phase control, a higher detection resolution and higher parameter identification performance can be obtained.

Description

A kind of high-resolution Underwater Target Detection device and method based on MIMO

Technical field

The present invention relates to a kind of high-resolution Underwater Target Detection method based on MIMO, is a kind of sonar signal New Method for Processing, belongs to Underwater Acoustic Object and detects and positioning field.

Background technology

Existing Underwater Target Detection method based on level transmitting-receiving battle array structure all is the phased array structure; The emission battle array is driven same detectable signal and is emitted to a certain direction; Mode through scanning is surveyed all interested directions; Receive battle array and receive signal echo, realize target detection and location through wave beam formation to received signal and matched filtering.The target resolution of phased array detection method becomes ratio with the battle array aperture that receives battle array.Therefore can penetrate the aperture of battle array and reach through increasing transmitting-receiving for the resolution that improves target, but the expansion in battle array aperture needs more emitting and receiving equipment, designs more complicated sonar system, pays bigger economic cost.And phased array be directive property emission with a kind of signal, once can only detect an interested orientation, realize that omnibearing detection need need a lot of detection times through repeatedly comprehensive scanning.

Summary of the invention

The objective of the invention is problem, a kind of high-resolution Underwater Target Detection method based on MIMO is provided to the existence of existing sonar technique.

MIMO refers to multiple-input and multiple-output; How input refers to is launched the quadrature detectable signal simultaneously, and many outputs refer to receive in gust signal echo that receives and comprise the target information that different detectable signals are taken back, and form and matched filter processing through wave beam; Distinguish the transmit virtual data vector of echo of difference; Obtain the waveform diversity gain, expand virtual battle array aperture, arrive high-resolution detection and location.MIMO is a non-directive emission orthogonal waveforms simultaneously, and once emission can realize the detection to all orientation, thereby can save more detection time.

For realizing above-mentioned purpose, the technical scheme that the present invention taked is: the high-resolution Underwater Target Detection device that the present invention is based on MIMO comprises:

Signal transmitter is used to launch orthogonal signal;

Power amplifier is used for orthogonal signal are carried out power amplification;

The transmitting transducer battle array is used for the orthogonal signal after amplifying are converted into acoustical signal by electric signal, and acoustical signal is emitted to desire detection waters;

The receiving hydrophone battle array is used to receive the echo of the acoustical signal that said transmitting transducer paroxysm penetrates, and converts the echo of said acoustical signal into electric signal;

Signal receiver is used for receiving and stores said receiving hydrophone battle array institute electrical signal converted;

Signal processor is used to receive the electric signal of receiving from signal receiver, and this electric signal is carried out MIMO handle, and obtains the MIMO beam modes; Utilize formula (1) that the MIMO beam modes is estimated, obtain the orientation of potential target, utilize the Generalized Likelihood Ratio detecting device shown in the formula (2) whether to detect on the orientation of said potential target the necessary being target then:

$\widehat{\mathrm{\θ}}=\arg \underset{\mathrm{\θ},\mathrm{\α}}{\min }Q\left(\mathrm{\θ}\right)---\left(1\right)$

$Q\left(\widehat{\mathrm{\&theta;}}\right)\begin{array}{c}\stackrel{H_1}{\&GreaterEqual;}\\ \underset{{\mathrm{<figure-callout\; id="0"\; label="H"\; filenames="111219112430.png,111219112437.png"\; state="\{\{state\}\}">H</figure-callout>}}_0}{<}\end{array}\mathrm{\&delta;}---\left(2\right)$

In formula (1) and (2), Q (θ) is the MIMO beam modes, and θ is all orientation that target possibly exist, and α is the reflection coefficient of target,

Be the orientation of the potential target that estimates,

Be the beam modes value on the potential target orientation that estimates, H ₀Represent not have real target, H on the orientation of the estimated potential target that goes out ₁Represent to have real target on the orientation of the estimated potential target that goes out, δ is the detection threshold of Generalized Likelihood Ratio detecting device.

The method of utilizing the high-resolution Underwater Target Detection device based on MIMO of the present invention to carry out the high-resolution Underwater Target Detection comprises the steps:

(1) signal transmitter sends one group of M mutually orthogonal orthogonal signal, and said orthogonal signal are expressed as s [n]=(s ₁[n], s ₂[n] ... S _M[n]) ^T,

And said orthogonal signal satisfy the signal in orthogonal condition shown in the formula (3):

$\frac{1}{\sqrt{L}}\underset{n=0}{\overset{L}{\mathrm{\&Sigma;}}}{s}_i\left[n\right]{s_j}^{*}\left[n\right]=\left\{\begin{array}{cc}1,& i=\mathrm{<figure-callout\; id="0"\; label="j"\; filenames="111219112430.png,111219112437.png"\; state="\{\{state\}\}">j</figure-callout>}\\ 0,& i\&NotEqual;j\end{array}\right.i,j=\mathrm{1,2}\&CenterDot;\&CenterDot;\&CenterDot;M---\left(3\right)$

Wherein, M is>=1 positive integer, and n representes n time-sampling point of said orthogonal signal, and 1≤n≤L, L are the length of orthogonal signal, the computing of T representing matrix transposition, s _iI among [n] said M the orthogonal signal s of expression [n], s _jJ among [n] said M the orthogonal signal s of expression [n], s _j ^*[n] expression is to s _jThe conjugate operation of [n];

(2) be positioned over desire to the transmitting transducer battle array that comprises M unit transmitting transducer with the receiving hydrophone battle array that comprises N unit receiving hydrophone and survey in the waters, wherein, N is>=1 positive integer; The spacing of adjacent transmitting transducer is d _t, the spacing of adjacent receiving hydrophone is d _r, the driving vector of the driving vector sum receiving hydrophone battle array of said transmitting transducer battle array is respectively suc as formula shown in (4), (5); After amplifying, M the orthogonal signal s [n] that said power amplifier is launched signal transmitter be sent to said transmitting transducer battle array; Orthogonal signal after said transmitting transducer battle array will be amplified are converted into acoustical signal by electric signal, and this acoustical signal is emitted to desire detection waters; Simultaneously said receiving hydrophone battle array receives the echo of the acoustical signal that the transmitting transducer battle array launched, and the echo of said acoustical signal is converted into suc as formula the electric signal shown in (6):

$a_t\left(\mathrm{\&theta;}\right)={[1,e^{-{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="111219112430.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;fd}}_t\sin \mathrm{\&theta;}/c},...,e^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="111219112430.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;f}(M-1)d_t\sin \mathrm{\&theta;}/c}]}^T---\left(4\right)$

$a_r\left(\mathrm{\&theta;}\right)={[1,e^{-{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="111219112430.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;fd}}_r\sin \mathrm{\&theta;}/c},...,e^{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="111219112430.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;f}(N-1)d_r\sin \mathrm{\&theta;}/c}]}^T---\left(5\right)$

In formula (4) and the formula (5), a _t(θ) be the driving vector of transmitting transducer battle array, a _rBe the driving vector of receiving hydrophone battle array (θ), M is the element number of array of said transmitting transducer battle array, and N is the element number of array of said receiving hydrophone battle array, and j is a complex unit, d _tBe the spacing of adjacent transmitting transducer, d _rBe the spacing of adjacent receiving hydrophone, f is the frequency of the orthogonal signal of signal transmitter emission, and c is the velocity of sound of water, and θ is all orientation that target possibly exist, the computing of T representing matrix transposition;

r[n]＝αa _r(θ)a _t ^T(θ)s[n]+w[n]，1≤n≤L (6)

In the formula (6), the electric signal that r [n] receives for the receiving hydrophone battle array, θ are the orientation at potential target place, and α is the reflection coefficient of target, and it is vectorial with the incoherent additive noise that transmits that w [n] receives for the receiving hydrophone battle array, a _t ^T(θ) expression is to a _tTransposition (θ);

(3) utilize electric signal r [n] that formula (7) receives the receiving hydrophone battle array respectively with each orthogonal signal s _i[n] does matched filter processing, obtain suc as formula the abundant statistical matrix shown in (8),

${\mathrm{\&eta;}}_i={\left[\begin{array}{cccc}{\mathrm{\&eta;}}_{i1}& {\mathrm{\&eta;}}_{i2}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&eta;}}_{\mathrm{iN}}\end{array}\right]}^T$

$\left(7\right)$

$=\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\left[n\right]{s_i}^{*}\left[n\right]={\mathrm{\&alpha;a}}_r\left(\mathrm{\&theta;}\right)a_{\mathrm{ti}}\left(\mathrm{\&theta;}\right),i=\mathrm{1,2},...M$

In the formula (7), s _iI among [n] said M the orthogonal signal s of expression [n], s _i ^*[n] expression is to s _iThe conjugate operation of [n]; η _In(n=1,2 ... N) be the vectorial η of abundant statistics of i unit _iN element, α _Ti(θ) (i=1,2 ... be that vectorial a is driven in emission M) _tI element (θ);

$E=\left[\begin{array}{cccc}{\mathrm{\&eta;}}_1& {\mathrm{\&eta;}}_2& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&eta;}}_M\end{array}\right]=\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\left[n\right]s^H\left[n\right]---\left(8\right)$

In the formula (8), E representes abundant statistical matrix, s ^H[n] expression is done the conjugate transpose computing to s [n];

Utilize formula (9) to carry out the column vector processing said abundant statistical matrix, be expanded suc as formula the vectorial η of abundant statistics shown in (10):

η＝vec([η ₁?η ₂?…?η _M]) (9)

$\mathrm{\&eta;}=\mathrm{vec}\left(\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\right[n\left]s^H\right[n\left]\right)$

$=\mathrm{vec}\left(\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}\left({\mathrm{\&alpha;a}}_r\left(\mathrm{\&theta;}\right){a_t}^T\left(\mathrm{\&theta;}\right)s\right[n]+w[n\left]\right)s^H\right[n\left]\right)---\left(10\right)$

$=\mathrm{\&alpha;vec}(a_r\left(\mathrm{\&theta;}\right){a_t}^T\left(\mathrm{\&theta;}\right)+\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}w[n\left]s^H\right[n\left]\right)$

$=\mathrm{\&alpha;d}\left(\mathrm{\&theta;}\right)+v$

In formula (9) and the formula (10), the computing of vec representing matrix column vector, s ^H[n] expression is conjugate transpose computing, a to s [n] _t ^T(θ) expression is to a _tTransposition (θ), v is distributed as v～N for obeying ^c(0, σ _w ²I _NM) multiple Gaussian noise, wherein, σ _w ²Be noise power, I _NMBe that order is the unit matrix of MN, d (θ)=vec (a _r(θ) a _t ^T(θ)) be the response vector of the receiving hydrophone battle array of length M N * 1;

The vectorial η of abundant statistics of the expansion that (4) obtains according to step (3); Obtain MIMO beam modes figure by formula (11), and obtain the maximal possibility estimation

in orientation, potential target place by formula (12)

$Q\left(\mathrm{\&theta;}\right)=\frac{{\left|d^H\left(\mathrm{\&theta;}\right)\mathrm{\&eta;}\right|}^2}{{\left|\right|d\left(\mathrm{\&theta;}\right)\left|\right|}^2}=\frac{{\left|{a_r}^H\left(\mathrm{\&theta;}\right)\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\right[n\left]s{\left[n\right]}^H{a_t}^{*}\left(\mathrm{\&theta;}\right)\right|}^2}{L{\left|\right|a_r\left(\mathrm{\&theta;}\right)\left|\right|}^2{\left|\right|a_t\left(\mathrm{\&theta;}\right)\left|\right|}^2}---\left(11\right)$

$\widehat{\mathrm{\&theta;}}=\arg \underset{\mathrm{\&theta;},\mathrm{\&alpha;}}{\min }Q\left(\mathrm{\&theta;}\right)---\left(12\right)$

In formula (11) and the formula (12), Q (θ) is the MIMO beam modes, and θ is all orientation that target possibly exist, a _t(θ) be the driving vector of transmitting transducer battle array, a _r(θ) be the driving vector of receiving hydrophone battle array, || || the expression vector is asked modular arithmetic, || for asking signed magnitude arithmetic(al), L is emission orthogonal signal length, a _r ^H(θ) expression is to a _r(θ) be transpose conjugate computing, a _t ^*(θ) expression is to a _t(θ) do conjugate operation, α representes the reflection coefficient of target,

The maximal possibility estimation in orientation, expression potential target place;

(5) judge according to formula (13) and whether locate the necessary being target in the orientation

of the estimated potential target that goes out of step (4):

$Q\left(\widehat{\mathrm{\&theta;}}\right)\begin{array}{c}\stackrel{H_1}{\&GreaterEqual;}\\ \underset{{\mathrm{<figure-callout\; id="0"\; label="H"\; filenames="111219112430.png,111219112437.png"\; state="\{\{state\}\}">H</figure-callout>}}_0}{<}\end{array}\mathrm{\&delta;}---\left(13\right)$

In the formula (13), Be the beam modes value on the potential target orientation that estimates, H ₀The orientation of representing the estimated potential target that goes out does not exist real target, H ₁There is real target in the orientation of representing the estimated potential target that goes out, and δ is the detection threshold of Generalized Likelihood Ratio detecting device.

Compared with prior art, the invention has the beneficial effects as follows:

1) handles through MIMO; The gain of acquisition waveform diversity has been expanded sonar virtual battle array aperture and spatial degrees of freedom, when the target localization wave beam forms; Narrower detection main lobe that can obtain and lower secondary lobe; Improve target detection resolution and parameter recognition performance, reduced the influence that secondary lobe disturbs, improved the echo signal to noise ratio (S/N ratio).

2) detection method that uses among the present invention is once surveyed the detection that can realize all orientation, has avoided the process of the comprehensive scanning of phased array pattern, can save detection time.

Description of drawings

Fig. 1 is the fundamental diagram of method among the present invention;

Fig. 2 is the synoptic diagram of 7 yuan of transmitting transducer battle arrays and 6 yuan of receiving hydrophone battle arrays;

Fig. 3 is the waveguide pool construction and the synoptic diagram of structuring the formation;

Fig. 4 is 7 tunnel orthogonal signal synoptic diagram that the transmitting transducer paroxysm is penetrated;

Fig. 5 is the beam pattern contrast of detection method and the phased array detection of a target among the present invention.

Embodiment

Below in conjunction with accompanying drawing and instantiation the present invention is done further description.

The present invention is based on the high-resolution Underwater Target Detection device of MIMO, comprising:

---signal transmitter is used to launch orthogonal signal;

---power amplifier is used for orthogonal signal are carried out power amplification;

---the transmitting transducer battle array is used for the orthogonal signal after amplifying are converted into acoustical signal by electric signal, and acoustical signal is emitted to desire detection waters;

---the receiving hydrophone battle array is used to receive the echo of the acoustical signal that said transmitting transducer paroxysm penetrates, and converts the echo of said acoustical signal into electric signal;

---signal receiver is used for receiving and stores said receiving hydrophone battle array institute electrical signal converted;

---signal processor, be used to receive the electric signal of receiving from signal receiver, and this electric signal carried out MIMO handle, obtain the MIMO beam modes; Utilize formula (1) that the MIMO beam modes is estimated, obtain the orientation of potential target, utilize the Generalized Likelihood Ratio detecting device shown in the formula (2) whether to detect on the orientation of said potential target the necessary being target then:

$\widehat{\mathrm{\&theta;}}=\arg \underset{\mathrm{\&theta;},\mathrm{\&alpha;}}{\min }Q\left(\mathrm{\&theta;}\right)---\left(12\right)$

$Q\left(\widehat{\mathrm{\&theta;}}\right)\begin{array}{c}\stackrel{H_1}{\&GreaterEqual;}\\ \underset{{\mathrm{<figure-callout\; id="0"\; label="H"\; filenames="111219112430.png,111219112437.png"\; state="\{\{state\}\}">H</figure-callout>}}_0}{<}\end{array}\mathrm{\&delta;}---\left(13\right)$

In formula (1) and (2), Q (θ) is the MIMO beam modes, and θ is all orientation that target possibly exist, and α is the reflection coefficient of target,

Be the orientation of the potential target that estimates,

Be the beam modes value on the potential target orientation that estimates, H ₀The orientation of representing said potential target does not exist target, H ₁There is target in the orientation of representing said potential target, and δ is the detection threshold of Generalized Likelihood Ratio detecting device.

Shown in Figure 1 for the fundamental diagram of detection method of the present invention.As shown in Figure 1; Sniffer of the present invention comprises above water and underwater portion; Above water comprises signal transmitter, power amplifier, signal receiver and signal processor; Underwater portion comprises transmitting transducer battle array (being called for short " emission battle array ") and receiving hydrophone battle array (being called for short " receiving battle array ") that comprises 6 yuan of receiving hydrophones of 7 yuan of transmitting transducers, and launch a burst of first number this moment is M=7, and receiving a burst of first number is N=6.

Need to prove; Though Fig. 1 is example for convenient explanation with 7 yuan of transmitting transducers and 6 yuan of receiving hydrophones; But the transmitting transducer in the transmitting transducer battle array of the present invention can be a M unit, and the receiving hydrophone in the receiving hydrophone battle array can be a N unit, wherein; The value of M is identical with the quantity of the mutually orthogonal orthogonal signal that signal transmitter is sent, and M is >=1 positive integer; N is the number that receives array element, and the number that receives array element is the positive integer greater than 1, and promptly N is >=1 positive integer.

The emission battle array is as shown in Figure 2 with the synoptic diagram of structuring the formation that receives battle array, and the battle array spacing of emission battle array is 10cm, and the battle array spacing that receives battle array is 6cm, is reference origin with first emission array element, and vectorial a is driven in emission so _t(θ) drive vectorial a with receiving _r(θ) can use the expression of formula (14) and formula (15) respectively,

a _t(θ)＝exp(-j2πd _tf?sin(θ)/c)，d _t＝[010?20?30?40?50?60] ^T/100；(14)

a _r(θ)＝exp(j2πd _rf?sin(θ)/c)，d _r＝[10?16?22?28?34?40] ^T/100；(15)

In formula (14) and the formula (15), f is an emission signal frequency, and c is the velocity of sound of water, and θ is all orientation that target possibly exist, the computing of T representing matrix transposition.

Because the pure-tone polse signal of different frequency can guarantee mutually orthogonal and satisfy the signal in orthogonal condition that formula (1) is represented, so the pure-tone polse signal that the present invention can use different frequency is as transmitting.It is that 1V, time span are the pure-tone polse signal s [n] of the different frequency of 3ms that signal transmitter shown in Figure 1 is launched 7 tunnel amplitudes, and 1≤n≤L, L are signal length.The expression formula of s [n] is suc as formula shown in (16).

$s\left(n\right)=\left(\begin{array}{c}{s}_1\left[n\right]\\ s_2\left[n\right]\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ s_M\left[n\right]\end{array}\right)=\left(\begin{array}{c}\sin ({2\mathrm{\&pi;f}}_{1}n/f_s)\\ \sin ({2\mathrm{\&pi;f}}_{2}n/f_s)\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ \sin ({2\mathrm{\&pi;f}}_{M}n/f_s)\end{array}\right),n=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;L---\left(16\right)$

S in the formula (16) _i[n] is the i road signal of 7 tunnel orthogonal signal s [n] of signal transmitter emission, f=[f ₁f ₂F _M]=[6,7,8,9,10,11,12] ^TKHz is respectively the frequency of the orthogonal signal of each road emission, f _s=50KHz is for receiving the signal sampling frequency, and then the 3ms signal length is the product of signal sampling frequency and time, that is: L=3*10 ^-3* 50*10 ³=150.Shown in Figure 3 is the 7 road quadratures (s that transmit ₁[n], s ₂[n] ... S _M[n]).

An emission battle array shown in Figure 2 and a reception battle array level are put into waveguide pond shown in Figure 4, and the long 8m in pond, wide 2m, dark 1.5m are that the sound eliminating tile wall is used for absorbing arrival pond sound wave all around around the pond, reduce multipath and disturb, the environment of simulation free field.Transmitting transducer battle array and receiving hydrophone battle array are positioned at depth of water 0.75m place, put a target apart from the θ=0 ° direction of this horizontal array 5m.

7 tunnel orthogonal signal s (n) of signal transmitter emission amplify through power amplifier; Be sent to the transmitting transducer battle array, the electrical signal conversion that transmitting transducer is sent out the power amplifier that receives big is that acoustical signal is emitted in the water, and No. 6 receiving hydrophone battle arrays receive the echo of the acoustical signal that said transmitting transducer paroxysm penetrates simultaneously; And convert the echo of said acoustical signal into electric signal; Signal receiver receives and stores said receiving hydrophone battle array institute electrical signal converted, the signal r (n) that this No. 6 receiving hydrophone battle array receives, 1≤n≤L; Expression is suc as formula shown in (17)

$r\left(n\right)=\left(\begin{array}{c}{r}_1\left[n\right]\\ r_2\left[n\right]\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ r_N\left[n\right]\end{array}\right),n=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;L,N=6---\left(17\right)$

R in the formula (17) _iThe signal that [n] expression i unit receiving hydrophone receives.

Utilize electric signal r [n] that formula (17) receives the receiving hydrophone battle array respectively with each orthogonal signal s _i[n] does matched filter processing, obtain suc as formula the abundant statistical matrix shown in (18),

${\mathrm{\&eta;}}_i={\left[\begin{array}{cccc}{\mathrm{\&eta;}}_{i1}& {\mathrm{\&eta;}}_{i2}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&eta;}}_{\mathrm{iN}}\end{array}\right]}^T$

$\left(18\right)$

$=\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\left[n\right]{s_i}^{*}\left[n\right]={\mathrm{\&alpha;a}}_r\left(\mathrm{\&theta;}\right)a_{\mathrm{ti}}\left(\mathrm{\&theta;}\right),i=\mathrm{1,2},...M$

In the formula (18), η _In(n=1,2 ... N) be the vectorial η of abundant statistics of i unit _iN element, a _Ti(θ) (i=1,2 ... be that vectorial a is driven in emission M) _tI element (θ);

$E=\left[\begin{array}{cccc}{\mathrm{\&eta;}}_1& {\mathrm{\&eta;}}_2& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&eta;}}_M\end{array}\right]=\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\left[n\right]s^H\left[n\right]---\left(19\right)$

E is abundant statistical matrix in the formula (19), s ^H[n] expression is done the conjugate transpose computing to s [n].

Utilize formula (19) to carry out the column vector processing said abundant statistical matrix, be expanded suc as formula the vectorial η of abundant statistics shown in (20):

η＝vec([η ₁?η ₂?…?η _M]) (19)

$\mathrm{\&eta;}=\mathrm{vec}\left(\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\right[n\left]s^H\right[n\left]\right)$

$=\mathrm{vec}\left(\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}\left({\mathrm{\&alpha;a}}_r\left(\mathrm{\&theta;}\right){a_t}^T\left(\mathrm{\&theta;}\right)s\right[n]+w[n\left]\right)s^H\right[n\left]\right)---\left(20\right)$

$=\mathrm{\&alpha;vec}(a_r\left(\mathrm{\&theta;}\right){a_t}^T\left(\mathrm{\&theta;}\right)+\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}w[n\left]s^H\right[n\left]\right)$

$=\mathrm{\&alpha;d}\left(\mathrm{\&theta;}\right)+v$

In formula (19) and the formula (20), the computing of vec representing matrix column vector, s ^H[n] expression is conjugate transpose computing, a to s [n] _t ^T(θ) expression is to a _t(θ) do the transposition computing, v is distributed as v～N for obeying ^c(0, σ _w ²I _NM) multiple Gaussian noise, σ wherein _w ²Be noise power, I _NMBe that order is the unit matrix of MN, d (θ)=vec (a _r(θ) a _t ^T(θ)) be the battle array response vector of the receiving hydrophone battle array of length M N * 1.And the element number of array that uses the receiving hydrophone battle array is when surveying as the conventional phased array of N; The length of the battle array response vector of its receiving hydrophone battle array is N * 1; The battle array response vector length of detection method of the present invention with respect to the extended length of the battle array response vector of phased array detection mode M doubly; Represent that virtual battle array aperture expanded M doubly, increased the additional space degree of freedom, thereby can improve the target resolution and the parameter recognition performance of detection system.

According to the vectorial η of abundant statistics that formula (20) obtains, set up the maximum likelihood ratio sensor of target azimuth θ.Obtain the maximal possibility estimation

of potential target orientation θ by formula (21)

$\widehat{\mathrm{\&theta;}}=\arg \underset{\mathrm{\&theta;},\mathrm{\&alpha;}}{\min }{\left|\right|\mathrm{\&eta;}-\mathrm{\&alpha;d}\left(\mathrm{\&theta;}\right)\left|\right|}^2$

$=\arg \underset{\mathrm{\&theta;},\mathrm{\&alpha;}}{\max }\frac{{\left|d^H\left(\mathrm{\&theta;}\right)\mathrm{\&eta;}\right|}^2}{{\left|\right|d\left(\mathrm{\&theta;}\right)\left|\right|}^2}---\left(21\right)$

In the formula (21) || || the expression vector is asked modular arithmetic, || for asking signed magnitude arithmetic(al).Can find out that by formula (21) maximal possibility estimation

of asking for the potential target orientation is equivalent to maximize the resulting θ value of objective function

.

Objective function in the formula (21)

simplification is written as Q (θ), shown in (22):

$Q\left(\mathrm{\&theta;}\right)=\frac{{\left|d^H\left(\mathrm{\&theta;}\right)\mathrm{\&eta;}\right|}^2}{{\left|\right|d\left(\mathrm{\&theta;}\right)\left|\right|}^2}=\frac{{\left|{a_r}^H\left(\mathrm{\&theta;}\right)\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\right[n\left]s^H\right[n\left]{a_t}^{*}\left(\mathrm{\&theta;}\right)\right|}^2}{L{\left|\right|a_r\left(\mathrm{\&theta;}\right)\left|\right|}^2{\left|\right|a_t\left(\mathrm{\&theta;}\right)\left|\right|}^2}---\left(22\right)$

In the formula (22), a is driven in emission _t(θ) drive vectorial a with receiving _rExpression formula (θ) is suc as formula shown in (14) and the formula (15), and the orthogonal signal s of emission [n] is suc as formula shown in (16), and the expression formula that receives signal is suc as formula shown in (17), and L=150 is the length of the orthogonal signal of signal transmitter emission.

The expressed beam modes of use formula (22) for each-90 °≤θ≤90 °, is calculated its corresponding beam modes value, obtains the beam modes figure shown in Fig. 5 solid line.According to formula (23) and beam modes figure; Search for the pairing angle θ of maximal value of Q (θ), be the orientation

of the potential target that estimates

Then according to the described Generalized Likelihood Ratio detecting device of formula (24), whether orientation

target of judging the potential target that estimates necessary being.

$Q\left(\widehat{\mathrm{\&theta;}}\right)\begin{array}{c}\stackrel{H_1}{\&GreaterEqual;}\\ \underset{{\mathrm{<figure-callout\; id="0"\; label="H"\; filenames="111219112430.png,111219112437.png"\; state="\{\{state\}\}">H</figure-callout>}}_0}{<}\end{array}\mathrm{\&delta;}---\left(24\right)$

In the formula (24), H ₀Represent not have real target, H on the orientation of the estimated potential target that goes out ₁Represent to have real target on the orientation of the estimated potential target that goes out, δ is the detection threshold of Generalized Likelihood Ratio detecting device, by predefined false-alarm probability P _fDecision.As Q (θ)>=P _fThe time think orientation, potential target place

The target necessary being, otherwise think orientation, potential target place

Target necessary being not.

Can find out that from Fig. 4 solid line the target azimuth

that estimation is come out is consistent with the realistic objective orientation, has proved the correctness of inventive method.

In order to contrast with the phased array situation; It is 3ms that each array element of emission battle array is launched a time span simultaneously; Amplitude is that the frequency band of 1V is the linear FM signal of 6Khz to 10Khz, and the signal of each array element emission of emission this moment battle array is identical, does not satisfy the signal in orthogonal condition shown in the formula (1); So be phased emission this moment, but not the MIMO emission.The reception signal that receives battle array is done the phased array wave beam form and matched filter processing, obtain its corresponding phased beam modes figure shown in Fig. 5 dotted line.

Based on Fig. 5, the beam modes figure that employed detection method in phased beam modes figure and the invention is obtained contrasts, and can find out that the beam modes output of detection method of the present invention can obtain narrower detection main lobe and lower detection secondary lobe.The main lobe width of the beam modes of detection method of the present invention is the first secondary lobe low 18dB of conventional phased array main lobe width

first secondary lobe than conventional phased array; Beam modes figure has narrower main lobe can obtain higher target resolution; Lower secondary lobe can reduce the interference of other directions; Improve the echo signal to noise ratio (S/N ratio), improve target detection probability.

Claims (2)

1. the high-resolution Underwater Target Detection device based on MIMO is characterized in that, comprising:

Signal transmitter is used to launch orthogonal signal;

Power amplifier is used for orthogonal signal are carried out power amplification;

The transmitting transducer battle array is used for the orthogonal signal after amplifying are converted into acoustical signal by electric signal, and acoustical signal is emitted to desire detection waters;

The receiving hydrophone battle array is used to receive the echo of the acoustical signal that said transmitting transducer paroxysm penetrates, and converts the echo of said acoustical signal into electric signal;

Signal receiver is used for receiving and stores said receiving hydrophone battle array institute electrical signal converted;

Signal processor is used to receive the electric signal of receiving from signal receiver, and this electric signal is carried out MIMO handle, and obtains the MIMO beam modes; Utilize formula (1) that the MIMO beam modes is estimated, obtain the orientation of potential target, utilize the Generalized Likelihood Ratio detecting device shown in the formula (2) whether to detect on the orientation of said potential target the necessary being target then:

$\widehat{\mathrm{\&theta;}}=\arg \underset{\mathrm{\&theta;},\mathrm{\&alpha;}}{\min }Q\left(\mathrm{\&theta;}\right)---\left(1\right)$

$Q\left(\widehat{\mathrm{\&theta;}}\right)\begin{array}{c}\stackrel{H_1}{\&GreaterEqual;}\\ \underset{H_0}{<}\end{array}\mathrm{\&delta;}---\left(2\right)$

In formula (1) and (2), Q (θ) is the MIMO beam modes, and θ is all orientation that target possibly exist, and α is the reflection coefficient of target, Be the orientation of the potential target that estimates,

Be the beam modes value on the potential target orientation that estimates, H ₀Represent not have real target, H on the orientation of the estimated potential target that goes out ₁Represent to have real target on the orientation of the estimated potential target that goes out, δ is the detection threshold of Generalized Likelihood Ratio detecting device.

2. a device that utilizes claim 1 carries out the method for high-resolution Underwater Target Detection, it is characterized in that, comprises the steps:

(1) signal transmitter sends one group of M mutually orthogonal orthogonal signal, and said orthogonal signal are expressed as s [n]=(s ₁[n], s ₂[n] ... S _M[n]) ^T,

And said orthogonal signal satisfy the signal in orthogonal condition shown in the formula (3):

$\frac{1}{\sqrt{L}}\underset{n=0}{\overset{L}{\mathrm{\&Sigma;}}}{s}_i\left[n\right]{s_j}^{*}\left[n\right]=\left\{\begin{array}{cc}1,& i=j\\ 0,& i\&NotEqual;j\end{array}\right.i,j=\mathrm{1,2}\&CenterDot;\&CenterDot;\&CenterDot;M---\left(3\right)$

Wherein, M is>=1 positive integer, and n representes n time-sampling point of said orthogonal signal, and 1≤n≤L, L are the length of orthogonal signal, the computing of T representing matrix transposition, s _iI among [n] said M the orthogonal signal s of expression [n], s _jJ among [n] said M the orthogonal signal s of expression [n], s _j ^*[n] expression is to s _jThe conjugate operation of [n].

(2) be positioned over desire to the transmitting transducer battle array that comprises M unit transmitting transducer with the receiving hydrophone battle array that comprises N unit receiving hydrophone and survey in the waters, wherein, N is>=1 positive integer; The spacing of adjacent transmitting transducer is d _t, the spacing of adjacent receiving hydrophone is d _r, the driving vector of the driving vector sum receiving hydrophone battle array of said transmitting transducer battle array is respectively suc as formula shown in (4), (5); After amplifying, M the orthogonal signal s [n] that said power amplifier is launched signal transmitter be sent to said transmitting transducer battle array; Orthogonal signal after said transmitting transducer battle array will be amplified are converted into acoustical signal by electric signal, and this acoustical signal is emitted to desire detection waters; Simultaneously said receiving hydrophone battle array receives the echo of the acoustical signal that the transmitting transducer battle array launched, and the echo of said acoustical signal is converted into suc as formula the electric signal shown in (6):

$a_t\left(\mathrm{\&theta;}\right)={[1,e^{-{j2\mathrm{\&pi;fd}}_t\sin \mathrm{\&theta;}/c},...,e^{-j2\mathrm{\&pi;f}(M-1)d_t\sin \mathrm{\&theta;}/c}]}^T---\left(4\right)$

$a_r\left(\mathrm{\&theta;}\right)={[1,e^{-{j2\mathrm{\&pi;fd}}_r\sin \mathrm{\&theta;}/c},...,e^{-j2\mathrm{\&pi;f}(N-1)d_r\sin \mathrm{\&theta;}/c}]}^T---\left(5\right)$

In formula (4) and the formula (5), a _t(θ) be the driving vector of transmitting transducer battle array, a _rBe the driving vector of receiving hydrophone battle array (θ), M is the element number of array of said transmitting transducer battle array, and N is the element number of array of said receiving hydrophone battle array, and j is a complex unit, d _tBe the spacing of adjacent transmitting transducer, d _rBe the spacing of adjacent receiving hydrophone, f is the frequency of the orthogonal signal of signal transmitter emission, and c is the velocity of sound of water, and θ is all orientation that target possibly exist, the computing of T representing matrix transposition;

r[n]＝αa _r(θ)a _t ^T(θ)s[n]+w[n]，l≤n≤L (6)

In the formula (6), the electric signal that r [n] receives for the receiving hydrophone battle array, θ are the orientation at potential target place, and α is the reflection coefficient of target, and it is vectorial with the incoherent additive noise that transmits that w [n] receives for the receiving hydrophone battle array, a _t ^T(θ) expression is to a _tTransposition (θ).

(3) utilize electric signal r [n] that formula (7) receives the receiving hydrophone battle array respectively with each orthogonal signal s _i[n] does matched filter processing, obtain suc as formula the abundant statistical matrix shown in (8),

${\mathrm{\&eta;}}_i={\left[\begin{array}{cccc}{\mathrm{\&eta;}}_{i1}& {\mathrm{\&eta;}}_{i2}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&eta;}}_{\mathrm{iN}}\end{array}\right]}^T$

$=\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\left[n\right]{s_i}^{*}\left[n\right]={\mathrm{\&alpha;a}}_r\left(\mathrm{\&theta;}\right)a_{\mathrm{ti}}\left(\mathrm{\&theta;}\right),i=\mathrm{1,2},...M---\left(7\right)$

In the formula (7), s _iI among [n] said M the orthogonal signal s of expression [n], s _i ^*[n] expression is to s _iThe conjugate operation of [n]; η _In(n=1,2 ... N) be the vectorial η of abundant statistics of i unit _iN element, α _Ti(θ) (i=1,2 ... be that vectorial a is driven in emission M) _tValency element (θ);

$E=\left[\begin{array}{cccc}{\mathrm{\&eta;}}_1& {\mathrm{\&eta;}}_2& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{\&eta;}}_M\end{array}\right]=\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\left[n\right]s^H\left[n\right]---\left(8\right)$

In the formula (8), E representes abundant statistical matrix, s ^H[n] expression is done the conjugate transpose computing to s [n];

Utilize formula (9) to carry out the column vector processing said abundant statistical matrix, be expanded suc as formula the vectorial η of abundant statistics shown in (10):

η＝vec([η ₁?η ₂?…?η _M]) (9)

$\mathrm{\&eta;}=\mathrm{vec}\left(\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\right[n\left]s^H\right[n\left]\right)$

$=\mathrm{vec}\left(\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}\left({\mathrm{\&alpha;a}}_r\left(\mathrm{\&theta;}\right){a_t}^T\left(\mathrm{\&theta;}\right)s\right[n]+w[n\left]\right)s^H\right[n\left]\right)---\left(10\right)$

$=\mathrm{\&alpha;vec}(a_r\left(\mathrm{\&theta;}\right){a_t}^T\left(\mathrm{\&theta;}\right)+\frac{1}{\sqrt{L}}\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}w[n\left]s^H\right[n\left]\right)$

$=\mathrm{\&alpha;d}\left(\mathrm{\&theta;}\right)+v$

In formula (9) and the formula (10), the computing of vec representing matrix column vector, s ^H[n] expression is conjugate transpose computing, a to s [n] _t ^T(θ) expression is to a _tTransposition (θ), v is distributed as v～N for obeying ^c(0, σ _w ²I _NM) multiple Gaussian noise, wherein, σ _w ²Be noise power, I _NMBe that order is the unit matrix of MN, d (θ)=vec (a _r(θ) a _t ^T(θ)) be the response vector of the receiving hydrophone battle array of length M N * 1.

The vectorial η of abundant statistics of the expansion that (4) obtains according to step (3); Obtain MIMO beam modes figure by formula (11), and obtain the maximal possibility estimation in orientation, potential target place by formula (12)

$Q\left(\mathrm{\&theta;}\right)=\frac{{\left|d^H\left(\mathrm{\&theta;}\right)\mathrm{\&eta;}\right|}^2}{{\left|\right|d\left(\mathrm{\&theta;}\right)\left|\right|}^2}=\frac{{\left|{a_r}^H\left(\mathrm{\&theta;}\right)\underset{n=1}{\overset{L}{\mathrm{\&Sigma;}}}r\right[n\left]s{\left[n\right]}^H{a_t}^{*}\left(\mathrm{\&theta;}\right)\right|}^2}{L{\left|\right|a_r\left(\mathrm{\&theta;}\right)\left|\right|}^2{\left|\right|a_t\left(\mathrm{\&theta;}\right)\left|\right|}^2}---\left(11\right)$

$\widehat{\mathrm{\&theta;}}=\arg \underset{\mathrm{\&theta;},\mathrm{\&alpha;}}{\min }Q\left(\mathrm{\&theta;}\right)---\left(12\right)$

In formula (11) and the formula (12), Q (θ) is the MIMO beam modes, and θ is all orientation that target possibly exist, a _t(θ) be the driving vector of transmitting transducer battle array, a _r(θ) be the driving vector of receiving hydrophone battle array, || || the expression vector is asked modular arithmetic, || for asking signed magnitude arithmetic(al), L is emission orthogonal signal length, a _r ^H(θ) expression is to a _r(θ) be transpose conjugate computing, a _t ^*(θ) expression is to a _t(θ) do conjugate operation, α representes the reflection coefficient of target,

The maximal possibility estimation in orientation, expression potential target place.

(5) judge according to formula (13) and whether locate the necessary being target in the orientation

of the estimated potential target that goes out of step (4):

$Q\left(\widehat{\mathrm{\&theta;}}\right)\begin{array}{c}\stackrel{H_1}{\&GreaterEqual;}\\ \underset{H_0}{<}\end{array}\mathrm{\&delta;}---\left(13\right)$

In the formula (13),

Be the beam modes value on the potential target orientation that estimates, H ₀The orientation of representing the estimated potential target that goes out does not exist real target, H ₁There is real target in the orientation of representing the estimated potential target that goes out, and δ is the detection threshold of Generalized Likelihood Ratio detecting device.

Images