CN1110701C - Method for processing sonar wave beams and its data processing system - Google Patents

Abstract

The present invention relates to a method for processing sonar wave beams and a data processing system thereof. A groove noise field method is adopted, and an array manifold alpha(that<i>) which is estimated from actually measured array data is introduced to determine an optimal weight coefficient vector W<OPT>. The array data processing system which realizes the wave beam optimization method is composed of a computer and a data acquisition card, wherein the data acquisition card is a high-speed array data acquisition card which is composed of a data buffer isolator, a handshake circuit, a CPU, a program memory, a data memory and a decoding circuit. The present invention solves the problem of optimization design of determining sensor array directional patterns in optional structural shapes by academic modeling, and has the advantages of reasonable system design and convenient use.

Description

A kind of sonar wave beams disposal route and data handling system thereof

Technical field: the present invention relates to a kind of sonar wave beams disposal route and data handling system thereof, be mainly used in the data acquisition of arbitrary shape sensor array under water and the optimal design of beam pattern.Sonar wave beams figure optimization method adopts " groove noise field method " and adaptive array treatment technology to finish, for the data handling system of supporting sonar wave beams disposal route proposed by the invention to design is made up of data acquisition and data processor.

Background technology: sonar wave beams has clear and definite requirement to main lobe shape, beam position error, beamwidth, secondary lobe level etc.For solving the optimal design of sonar wave beams figure, desirable modeling optimization method was proposed once in theory.As " groove noise field method ", be a kind of pure theory modeling beam optimization method.This method has proposed the notion of " groove noise field method " and " design of arbitrary shape sensor array directional diagram optimum optimization ", and its main points have three: all primitives of (1), composition basic matrix, its roomage response is a uniformity.This only is theoretic desirable imagination, and may there be the roomage response of uniformity in the actual conditions on the engineering; (2), the main lobe direction of actual beam, and make that its noise intensity is zero, all the other directions are even noise intensity, this is a groove noise field method.This also is desirable modeling; (3) signal response that will retrain the wave beam main lobe direction is made as 1, also is with the Utopian processing of engineering problem.Then, the adaptive array treatment theory is applied to the sensor basic matrix of arbitrary shape.But, the result who relies on idealized modeling to obtain, when it was applied to actual sonar system, measured beam pattern quality was not good.This is because theoretical and in esse distance.In practical engineering application, because complicated formation structure can cause that sound covers, scattering and transmission effect, and the reason of manufacturing process can cause the shape and size error of basic matrix, makes that array element response (array manifold) is inconsistent; Everything makes the beam pattern can't be met actual needs, the phenomenon of measured beam pattern performance substantial deviation expected results occurred.So in engineering is used, wish to find a kind of new weight coefficient optimization method, always so that all can obtain practical beam pattern to the sensor array of the arbitrary shape of any reality.

For fear of the deficiencies in the prior art part, the present invention proposes a kind of sonar wave beams disposal route and data handling system thereof.This method is based upon on the actual measurement array data basis fully based on the adaptive beam-forming algorithm of array data restructuring array stream shape.The array data disposal system is a kind of integrated digital display circuit of function that is independent of signal processor, not only can realize array data collection and beam pattern optimal design, and can do beam pattern and measure.

Summary of the invention: thought of the present invention is to solve optimal design and the measurement of the random geometry sensor array beam figure in the actual engineering.Consideration is for the sensor array of a given arbitrary structures shape, adopts " groove noise field method ", keeping under main lobe direction and beamwidth is certain and output noise power the is minimum condition, and the application self-adapting wave beam forms theory and finds the solution-organize the best weights coefficient.The setting array signal phasor is C, and noise vector is X, and W is the best weights coefficient, tries to achieve beamformer output signal Y=W ^T(X+C); It is characterized in that: the best weights coefficient amount of owing W _OPTAsk for not

It is resulting to be that theory is found the solution again $W_{\mathrm{opt}}=\frac{R_{\mathrm{XX}}^{-1}{C}^{*}}{C^T{R}_{\mathrm{XX}}^{-1}{C}^{*}},$ But introduce the array manifold α (θ that estimates from the actual measurement array data _i) back and the optimum weight vector W that tries to achieve _OPTThe optimum weight vector W that the present invention proposes _OPTTry to achieve by following method: at first groove noise field vector is introduced actual measurement array manifold α (θ _i) revised, and establish by J respectively from θ ₁, θ ₂, θ ₃... θ _JThe zero-mean narrow band noise σ of direction _iN _i(t), then $X\left(t\right)=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i\&CenterDot;N_i\left(t\right)\&CenterDot;\mathrm{\&alpha;}\left({\mathrm{\&theta;}}_i\right),$ Then according to adaptive modeling beam optimization method desired output signal $d\left(t\right)=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i\&CenterDot;N_i\left(t\right)\&CenterDot;D\left({\mathrm{\&theta;}}_i\right);$ Promptly get the covariance matrix of array data $R=E[X\left(t\right)\&CenterDot;X^H\left(t\right)]=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^2\&CenterDot;\mathrm{\&alpha;}\left({\mathrm{\&theta;}}_i\right)\&CenterDot;{\mathrm{\&alpha;}}^H\left({\mathrm{\&theta;}}_i\right),$ The covariance matrix of data and wanted signal $P=E[d^{*}\left(t\right)\&CenterDot;X\left(t\right)]=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^2\&CenterDot;D^{*}\left({\mathrm{\&theta;}}_i\right)\&CenterDot;\mathrm{\&alpha;}\left({\mathrm{\&theta;}}_i\right),$ Try to achieve optimum weight vector W according to the covariance matrix of above-mentioned array data and the covariance matrix of data and wanted signal _Opt=R ^-1P.Because the effect of having implied objective factor in the array manifold, so optimum weight vector W _OPTPlayed the caused deviation of the revised theory value of finding the solution.This improved beam optimization method can obtain beam pattern desirable, that satisfy actual needs.

The present invention proposes a kind of array data disposal system that realizes above-mentioned beam optimization method, is made up of computing machine 2 and data collecting card 1, can finish array data collection, array beams figure optimal design and array beams figure and measure.It is characterized in that: data collecting card 1 is a kind of high-speed array data collecting card, is made up of data buffering isolator 3, handshake circuit 4, CPU5, program storage 6 and data-carrier store 7 and decoding scheme 8.Handshake circuit 4 receives the handshake of sonars, and CPU5 control data buffer isolator 3 reads in data from sonar, after CPU5 searches synchronous head, just by the sensor sequence number array data is deposited in the data-carrier store 7.When a collection of data acquisition finishes, again the content in the data-carrier store 7 is deposited in the hard disc of computing machine 2.Program storage 6 is placed the executive routine of CPU5, the chip selection signal of decoding scheme 8 generating routine storeies 6 and data-carrier store 7.

Description of drawings:

Fig. 1: array data disposal system theory diagram

Fig. 2: data collecting card theory diagram

Fig. 3: embodiment block diagram 1-data collecting card 2-computing machine 3-data buffering isolator 4-handshake circuit 5-CPU 6-data-carrier store 7-program storage 8-decoding scheme 9-volume array 10-Signal Pretreatment 11-concentric cable 12-demodulating equipment

Embodiment: now in conjunction with the embodiments the present invention is further described:

Present embodiment is to utilize a kind of array data disposal system of the present invention, measures with beam pattern at the beam optimization design that a volume array of being made up of 64 array elements is finished.The computing machine 2 of array data disposal system is selected commercial PC for use, the data buffering isolator 3 of data collecting card 1 adopts ternary output bus to transmit receiver, handshake circuit 4 adopts monostalbe trigger, CPU5 adopts the high-speed dsp chip, program storage 6 adopts high-speed RAM, data-carrier store 7 adopts multiport memory, and decoding scheme 8 adopts the GAL circuit.

Front end in system, signal source are the volume arrays 9 of one 64 array element composition, and array signal is through sending into demodulating equipment 12 with the Signal Pretreatment 10 and the concentric cable 11 of volume array 9 connections.The signal of demodulation is sent into the array data disposal system.The array data disposal system is estimated array manifold α (θ according to the array data of the reality of gathering _i), calculate R and P, try to achieve optimum weight vector W _OPT, basis again $Y=W_{\mathrm{opt}}^T\&CenterDot;X+C$ Calculate beamformer output signal, thereby survey and draw out the panorama multi-beam figure of volume array.

Beneficial effect: the present invention's advantage compared to existing technology is:

Estimate the beam optimization method of array manifold based on array data, beam pattern this difficult problem that seriously degenerates when having solved optimization power that the pure theory modeling finds the solution and being applied to actual sonar system.The inventive method can be applied to the optimal design of arbitrary structures shape sensor array pattern, and this is that additive method can't be finished.Be the array data disposal system that realizes that this method designs, can finish the optimal design and the engineering beam pattern of complex array beam pattern and measure.System design is reasonable, and is easy to use.

Claims (3)

1, a kind of sonar wave beams disposal route, optimal design and measurement for random geometry sensor array beam figure, consideration is for the sensor array of a given arbitrary structures shape, adopt " groove noise field method ", keeping under main lobe direction and beamwidth is certain and output noise power the is minimum condition, the application self-adapting wave beam forms theory and finds the solution one group of best weights coefficient, the setting array signal phasor is C, noise vector is X, and W is the best weights coefficient, tries to achieve beamformer output signal Y=W ^T(X+C); It is characterized in that: optimum weight vector W _OPTAsk for that no longer to be that theory is found the solution resulting $W_{\mathrm{opt}}=\frac{R_{\mathrm{XX}}^{-1}{C}^{*}}{C^T{R}_{\mathrm{XX}}^{-1}{C}^{*}},$ But introduce the array manifold α (θ that estimates from the actual measurement array data _i) back and the optimum weight vector W that tries to achieve _OPTOptimum weight vector W _OPTTry to achieve by following method: at first groove noise field vector is introduced actual measurement array manifold α (θ _i) revised, and establish by J respectively from θ ₁, θ ₂, θ ₃... θ _JThe zero-mean narrow band noise σ of direction _iN _i(t), then $X\left(t\right)=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i\&CenterDot;N_i\left(t\right)\&CenterDot;\mathrm{\&alpha;}\left({\mathrm{\&theta;}}_i\right),$ Then according to adaptive modeling beam optimization method desired output signal $d\left(t\right)=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i\&CenterDot;N_i\left(t\right)\&CenterDot;D\left({\mathrm{\&theta;}}_i\right);$ Promptly get the covariance matrix of array data $R=E[X\left(t\right)\&CenterDot;X^H\left(t\right)]=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^2\&CenterDot;\mathrm{\&alpha;}\left({\mathrm{\&theta;}}_i\right)\&CenterDot;{\mathrm{\&alpha;}}^H\left({\mathrm{\&theta;}}_i\right);$ The covariance matrix of data and wanted signal $P=E[d^{*}\left(t\right)\&CenterDot;X\left(t\right)]=\underset{i=1}{\overset{J}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^2\&CenterDot;D^{*}\left({\mathrm{\&theta;}}_i\right)\&CenterDot;\mathrm{\&alpha;}\left({\mathrm{\&theta;}}_i\right);$ Try to achieve optimum weight vector according to the covariance matrix of above-mentioned array data and the covariance matrix of data and wanted signal

W _OPT＝R ^-1·P。

2, a kind of data handling system that realizes the sonar wave beams disposal route of claim 1, form by computing machine (2) and data collecting card (1), it is characterized in that: data collecting card (1) is a kind of high-speed array data collecting card, it consists of the handshake circuit (4) of accepting the sonar synchronizing signal and connects CPU (5), and the data buffering isolator (3) of preposition receiving array data, program storage (6), data-carrier store (7) and decoding scheme (8) connect on the bus of CPU (5).

3, a kind of data handling system as claimed in claim 2, it is characterized in that: the data buffering isolator (3) of data collecting card (1) adopts ternary output bus to transmit receiver, handshake circuit (4) adopts monostalbe trigger, CPU (5) adopts the high-speed dsp chip, program storage (6) adopts high-speed RAM, data-carrier store (7) adopts multiport memory, and decoding scheme (8) adopts the GAL circuit.

Images