CN100373808C - Signal processing system and method based on estimating signal direction of anived wave - Google Patents

Abstract

The present invention discloses a signal processing system of an array antenna, which realizes the estimate of signal direction of arrival (DOA) under the condition that the number of receivers is less than the number of antenna array elements. The system comprises an array antenna part, a switching matrix part, a signal receiver part, a control part, a covariance calculating part, a sub-array composition part (adder), a data memory part and a space feature detecting part, wherein antenna array elements (from 1-1 to 1-N) of the array antenna part induce and receive far-field arrival signals; the switching matrix part is used for connecting or disconnecting nodes according to the instruction of a control circuit; the signal receiver part is used for magnifying or attenuating the amplitude of output signals of the antenna array elements, carrying out AGC treatment to, demodulating and decoding the output signals of the antenna array elements so as to transform the arrival signals into digital baseband signals; the control part is used for controlling the connection or the disconnection of a switching matrix and the memory operation of a data memory; the covariance calculating part is used for calculating the covariance of the output signals of the array elements; the sub-array composition part (adder) is used for composing equivalent covariance; the data memory part is used for storing an intermediate result and a final result of data operation; the space feature detecting part is used for detecting the space feature parameter of the arrival signals to estimate the direction of arrival.

Description

Based on arrived wave signal direction estimated signals treatment system and method

Technical field

The present invention relates to satellite communication field, relate in particular to the processing unit and the method for the DOA estimation of signal in a kind of satellite communication, this method can be applicable to because system complexity and power consumption constraints processor number are less than the occasion of bay number.

Background technology

In spectrum is estimated, typically compose method of estimation and can be divided into three kinds: the period map method (periodogram) that Schuster in 1898 proposes is based on the direct spectrum method of estimation of Fourier conversion; Nineteen twenty-seven Yule propose based on AR, the model parameter The Power Spectrum Estimation Method of MA and arma modeling; The super-resolution spectrum method of estimation of Schmidt proposition in 1986, the MUSIC method is typical case's representative, is to use one of more method.MUSIC super-resolution spectrum method of estimation is based on the arrived wave signal Subspace Decomposition, and is closely related with array antenna array element spatial distribution characteristic (array manifold).The ESPRIT algorithm that people such as Roy propose also belongs to the super-resolution spectrum algorithm for estimating that decomposes based on proper subspace, this method has been utilized the symmetry of array antenna array element in spatial distribution, require the received signal vector space to have rotational invariance on the invariant subspace of spatial domain, the MUSIC method distributes to array element does not have the symmetry requirement.

DOA method of estimation based on the characteristic vector Subspace Decomposition is: the additive noise in arrived wave signal and array element adds under the mutual incoherent condition of interference signal, the full signal space of array antenna output signal can be decomposed into the subspace of two quadratures, the noise subspace that signal subspace that signal phasor is dominant and noise vector are dominant.The projection of vector in the signal subspace on noise subspace is zero vector.So the inner product of signal phasor and noise vector equals zero, it is infinitely great that its inverse is tending towards.Based on this, search just can obtain arrived wave signal direction (DOA) angle with the pairing wave angle that reaches of the null vector of noise vector inner product.

In actual applications, the data that we can observe are data of signal and noise stack, can be expressed as the output random data sequence x of array antenna _i(n), i=1,2 ..., M; N=1,2 ..., N.Wherein i represents the locus numbering of array antenna array element with respect to reference array element, and M is an array number; N represents the snap sampling constantly, and N is fast umber of beats.Be without loss of generality, can suppose that array dateout random sequence is that the interior broad sense of spatial domain and time-domain is steady, the random process of traversal.At M〉K, during N → ∞, can obtain array antenna output signal sequence x _i(n) ensemble average value, i.e. the valuation of covariance matrix,

R wherein _IjBe array antenna i, the assessment of covariation of j delayed output signals, K are that information source reaches wave number.Carry out DOA and estimate, need calculate the valuation of covariance matrix earlier

Each information source reach ripple be on spatial domain and the time-domain broad sense steadily, during ergodic random process, the ensemble average covariance matrix

Be equivalent to the statistical average covariance matrix of delayed output signals in the spatial domain.Angular resolution based on Subspace Decomposition DOA algorithm for estimating depends on the covariance matrix valuation

Levels of precision.Valuation

More little with the error of assembly average R, the angular resolution that DOA estimates is high more.The covariance matrix valuation of array output signal Calculating be to carry out the important content that DOA estimates.

In the DOA algorithm for estimating, according to different receiving system structures, the computational methods of covariance matrix can be divided into two classes, a class is based on the multi-receiver system configuration, and second class is based on the single-receiver system configuration.In existing system based on multi-receiver, each array element of array antenna is corresponding one by one with receiver (signal processor): logically, the array element output is directly connected to receiver inlet, in each snap sampling constantly, receiver needs to receive, handle the data of whole array element outputs, therefore, the quantity of receiver (signal processor) equates with bay quantity.Because the quantity of system receiver equals bay quantity, in the bigger application scenario of bay number, the corresponding increase of receiver quantity, on the one hand, system cost increases, and on the other hand, the total power consumption of receiver (signal processor) will increase rapidly.This type systematic can not be used in equipment power dissipation, the strict limited spaceborne communication system of weight and volume.Second type systematic has only a receiver, the output of each array element of array antenna links to each other with the phase shifter/delayer of a series of complex value weight coefficients, each phase shifter/delayer output links to each other with the adder input, and the output of adder links to each other with receiver (signal processor) input.The hardware cost of this system is lower, is equivalent to a time domain serial received system.Its shortcoming is when the bay number is bigger, and required weighting phase shift phase device and delayer quantity also increase, and calculates complex value amplitude and the required operand of phase weighting coefficient and increases sharply, and is unfavorable for the data in real time processing.These two kinds of systems all can not satisfy the requirement of satellite communication, are not suitable for using in satellite communication.

Summary of the invention

Based on the problems referred to above, be that common array signal processing system requires array antenna array element quantity to equal the quantity of signal receiver usually, and in some applications, perhaps can't satisfy this condition, and perhaps needing to make the two unequal---receiver quantity is less than array element quantity.The present invention discloses a kind of signal processing system, a kind of ripple DOA method of estimation and a kind of best subarray quantity computation method of reaching.

In satellite communication, spaceborne receiver quantity is lacked than array antenna array number M.Estimate resolution in order to improve DOA, must make full use of the array antenna resource, make its effective aperture reach maximum.This requirement must all use the ripple snap that reaches on whole array elements of antenna.On the other hand, in each snap sampling constantly, L _rIndividual receiver at most can only receiving array antenna L _rOutput snap on the individual array element.

Therefore, an object of the present invention is to provide a kind of array signal processing system, according in the array signal processing system of the present invention, a kind of time division multiplexing receiver structure scheme is disclosed, operation principle is as follows: different with traditional reception programme, in this reception structure, array antenna is not directly to link to each other with signal receiver, but links to each other with receiver by a programme-controlled exchange matrix.The programme-controlled exchange matrix is equivalent to electronic switch, and each node has connection and disconnects two operating states.In any snap sampling constantly, the programme-controlled exchange matrix makes always has L in the array antenna _rIndividual array element and L _rThe input of individual receiver links to each other; In different snap samplings constantly, the programme-controlled exchange matrix may switch connected node, makes another group L of receiver and array antenna _rIndividual array element links to each other.In the time of need switching, the L in the programme-controlled exchange matrix _rIndividual node switches simultaneously.Represent the programme-controlled exchange battle array with sw, then have:

Line display receiver numbering in formula (1) the sw matrix, tabulation are shown the array element numbering of main array antenna, k _{I, j}Expression node connection status, k _{I, j}∈ (0,1), k _{I, j}=0 expression does not connect k _{I, j}=1 expression receiver i links to each other with bay j, and then array element is exported fast beat of data and is input in the receiver.i＝1，2，…L _r，j＝0，1，…，L-1。

In foundation array signal processing system of the present invention, each array element of antenna realizes that with receiver grouping, timesharing are connected, and each snap sampling always has L constantly _rIndividual array element links to each other with receiver; Receiver constantly can receive L in each snap sampling _rThe fast beat of data of the output of individual array element, receiver is in running order always.

According in the array signal processing system of the present invention, can receive the fast beat of data of output of any amount bay with a small amount of receiver.Adopt the working method of time division multiplexing receiver, improved the utilance of receiver, reduced the quantity of required receiver, make the total power consumption of receiver controlled.

According in the array signal processing system of the present invention, signal processing method can both be connected all array elements of main array antenna with receiver, and the array element resource is fully used, thereby effective aperture that can the expanded matrix antenna, improves the DOA estimated resolution.

According in the array signal processing system of the present invention, the dateout of receiver structure on can the whole array elements of receiving array antenna obtains the full detail of arrived wave signal, reaches ripple DOA estimated resolution and reaches maximum thereby can make.What but scheme adopted is the mode of time division multiplexing receiver, and receiver receives only L constantly in each snap sampling _rThe output snap of individual array element.Array antenna is actual to be made up of L array element, the bay sum is greater than the receiver sum, therefore receiver can not receive data on the whole array elements of main array antenna constantly a snap sampling, therefore, can not directly calculate the covariance matrix of array antenna output snap in a conventional manner.For this reason, the present invention disclose a kind of array signal processing method-based on subarray synthetic reach the ripple direction determining method.

The receiver of signal processing system involved in the present invention (signal processor) quantity is between the receiver quantity of single-receiver and multi-receiver system, promptly greater than 1, less than array antenna array number (M), concrete quantity is according to applied environment condition and application requirements adjustment.

According in the array signal processing system of the present invention, to the array antenna type without limits, both can be uniform line-array, also can be the linear array of irregular alignment, or the array of plane distribution.

According in the array signal processing system of the present invention, the quantity of bay not to be strict with, array antenna is minimum can to comprise 2 array elements, and array element quantity for example reaches the wave number amount, processing speed etc. by the practical application conditional decision.

According in the array signal processing system of the present invention, the relative populations of signal receiver and bay is not had strict the restriction, but receiver quantity generally is less than bay quantity.When equaling bay quantity, receiver quantity just has been degenerated to common array signal processing system.

First aspect of the present invention is a kind of array antenna signal processing system, comprise: a kind of treatment system and corresponding processing method of estimating based on the synthetic DOA of subarray, the arrived wave signal direction (DOA) that this treatment system realization receiver quantity is less than under the bay quantity situation is estimated.This treatment system comprises: the array antenna part, and bay (1-1 to 1-N) induction receives the far field arrived wave signal; The switching matrix part is according to the instruction connection or the disconnected node of control circuit; The signal receiver part is amplified or decay bay amplitude output signal, carries out automatic gain control (AGC) and handles, and carries out demodulation, decoding is transformed to digital baseband signal with arrived wave signal; Control section, the connection or the disconnection of control switching matrix, the accessing operation of control data memory; The covariance calculating section, the covariance of calculating delayed output signals; Subarray composition part (adder), synthetic equivalent covariance; Data storage part storing intermediate result of data operation and final result; The space characteristics test section, the space characteristic parameter of detection arrived wave signal is realized reaching ripple direction (DOA) and is estimated.

Another aspect of the present invention provides a kind of signal processing method, and this method realizes reaching the super-resolution estimation of ripple direction (DOA) based on the arrived wave signal snap.Described method comprises: based on the time division multiplexing processing method, ask the covariance matrix of sub antenna array output; Ask the equivalent covariance matrix of main antenna array output; Reaching ripple DOA estimates.

The 3rd aspect of the present invention discloses to be optimized criterion, method, the scheme of choosing sub-array antenna and counts computing formula.

According to the present invention, can realize in the communication system, signal processing system of bay number that the super-resolution of arrived wave signal direction (DOA) estimates because system complexity and power consumption constraints, signal receiver (signal processor) number are less than.

System involved in the present invention can be applicable to a plurality of occasions, comprising: satellite communication system, onboard radar system, multiple-input and multiple-output (MIMO) mobile communication system, seismic signal detection system, subterranean resource exploration etc.

In foundation array signal processing system of the present invention, main array antenna 1-1 to 1-L is divided into several subarray antennas, the array number of each subarray antenna is less than the array number of main array antenna, and the array number of each subarray antenna equals the quantity of signal receiver.The array element of subarray antenna does not directly link to each other with receiver, but by switching matrix, adopts time division multiplexing mode to be connected with the input of receiver, realizes the time division multiplexing of receiver with this.The array number of each subarray equates.

In according to array signal processing system of the present invention and respective handling method, according to the array element numbering, main array antenna is divided into M sub-array antenna, each subarray antenna is by L _rIndividual array element is formed, and equals receiver quantity.The principle of dividing is: (1) exports fast beat of data by whole subarrays can synthesize the array antenna output correlation matrix of exporting the correlation matrix equivalence with main array, is called equivalent main array output correlation matrix; This requires any two array elements of main array antenna to occur once simultaneously in some subarray antennas at least.(2) minimum number of subarray antenna.

In according to array signal processing system of the present invention and respective handling method, because receiver quantity L _rBe less than main array antenna array number L, receiver constantly can only be received L in each snap sampling _rThe output snap of individual array element can't receive all the other L-L constantly in same sampling _rOutput snap on the individual array element.So, can not obtain to calculate the output snap that main array antenna is exported the required whole array elements of covariance matrix constantly a snap sampling.Yet, when arrived wave signal and the characteristic of channel satisfy certain condition, can obtain to calculate the required whole snaps of covariance by the mode of time-division processing.

In according to array signal processing system of the present invention and respective handling method, it is as follows to propose processing method at this situation: (1) is in each snap sampling moment, to the L of a sub-array antenna _rDateout on the individual array element is carried out synchronized sampling, obtains L _rThe output of individual array element reach wave datum; (2) constantly sampling is implemented in the output of different subarray bays in different snap samplings, through obtaining the snap of whole subarray bay outputs after several snap samplings constantly.(3) according to the signal processing method among the present invention the whole fast beat of data that receives is handled, can be calculated required covariance matrix valuation.

In foundation array signal processing system of the present invention, receiver is implemented the data of the whole array elements outputs of sub-array antenna constantly in each snap sampling and is taken a sample.By time-sharing multiplex, can receive the fast beat of data of output of handling any amount bay, realize reception and processing to a large amount of delayed output signals.Compare with other scheme, under certain condition, this programme can significantly improve systematic function, improves the precision that reaches ripple DOA resolving power and spectrum estimation, reduces the main lobe width that the digital wave number form becomes.

Description of drawings

Below in conjunction with accompanying drawing the present invention is illustrated in further detail:

Fig. 1 is according to array antenna signal processing system function composition frame chart of the present invention;

Fig. 2-Fig. 5 is the DOA estimation curve under different condition, and the estimation curve parameter among each figure is respectively:

Fig. 2: it is 15,45 ,-45 degree that fast umber of beats N=40, signal to noise ratio snr=20 dB, ripple reach the angle.

Fig. 3: it is 15,19 ,-60 degree that fast umber of beats N=400, signal to noise ratio snr=20 dB, ripple reach the angle.

Fig. 4: it is 60,64 ,-30 degree that fast umber of beats N=50, signal to noise ratio snr=100 dB, ripple reach the angle.

Fig. 5: it is 60,64 ,-30 degree that fast umber of beats N=10000, signal to noise ratio snr=30 dB, ripple reach the angle.

Embodiment

Below with reference to accompanying drawing of the present invention, more detailed description goes out most preferred embodiment of the present invention.

Fig. 1 forms structured flowchart according to the array antenna signal processing system function of the specific embodiment of the invention.Array antenna signal processing system shown in Figure 1 can be installed on the satellite, also can be installed in other array signal processing system, for example in the mobile communication equipment.

Explain the specific embodiment of the invention with reference to Fig. 1 below:

Array antenna in the signal processing system involved in the present invention (1) receives the signal that sends from one or more far fields equipment (for example, ground satellite station, base station etc.); The on off operating mode of control section (4) control switching matrix (2) node; Signal receiver (3) receives the arrived wave signal of switching matrix (2) output, becomes digital baseband signal after treatment; Covariance calculating section (5) calculates the estimated value of arrived wave signal correlation matrix; Subarray composition part (6) is based on the synthetic equivalent main array covariance matrix of valuation of subarray output covariance matrix, and the result is stored in the data storage (7); Wave angle is estimated to reach based on the covariance matrix valuation in space characteristics test section (8), realizes that super-resolution reaches ripple direction (DOA) and estimates.Specific implementation method is as follows:

Among Fig. 1, array antenna part (1) comprises bay 1-1 to 1-L, and L is greater than 1.In foundation array signal processing system of the present invention, the array antenna of being made up of whole array element 1-1 to 1-L is defined as main array antenna; The array antenna of being made up of part array element is defined as the subarray antenna.As required, different subarray antennas can comprise identical array element.

In according to array signal processing system of the present invention and corresponding signal process method, the quantity of basis signal receiver, main array antenna is divided into several subarray antennas, and each subarray antenna comprises the array element of equal number, and array element quantity equals signal receiver quantity.

In according to array signal processing system of the present invention and corresponding signal process method, according to sub-array partition criterion disclosed by the invention main array antenna array element to be divided into groups, each group array element is formed a sub-array antenna.Suppose that each subarray antenna is by L _rIndividual array element is formed, and promptly hypothesis has L _rIndividual receiver, 2≤L _r≤ L.The subarray antenna is serial number successively.

In foundation array signal processing system of the present invention, array antenna part (1) receives the wireless signal that comes from far-end, and the arrived wave signal complex amplitude of each array element output has different phase places.The signal of array element 1-1 to 1-N output becomes the input signal 2-i1 to 2-iL of switching matrix (2).Signal expression is as follows: be without loss of generality, establish: array antenna (1) is a uniform line-array, adjacent array element distance $d=\frac{\mathrm{\λ}}{2},$ λ is an arrived wave signal centre wavelength; Each subarray is by L _rIndividual array element is formed; K arrived wave signal, far field arrived wave signal complex amplitude is S _k(n), the angle of pitch is θ _k, uncorrelated mutually between signal and the signal; N (n) is an additive noise, obeys multiple Gaussian Profile, and average is zero, and variance is σ ², uncorrelated mutually between the noise on each array element, uncorrelated mutually between noise and the arrived wave signal.Then array element l at the arrived wave signal complex amplitude of sampling n induction constantly is

$y_l\left(n\right)=\underset{k=1}{\overset{k}{\mathrm{\&Sigma;}}}{S}_k\left(n\right)\exp (-\frac{\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}d\sin {\mathrm{\&theta;}}_k}{\mathrm{\&lambda;}})---\left(2\right)$

k＝1，2，...，K，l＝1，2，...，L。y _l(n) n corresponding to l array element exports snap.All L array element at the arrived wave signal complex amplitude vector of sampling n induction constantly is

Y(n) ^T＝[y ₁(n)y ₂(n)…y _L(n)]+N(n) (3)

Y (n) ^TCorresponding to the output snap of array antenna at sampling moment n.

Order $v_k=\exp \left(\frac{-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;}d\sin {\mathrm{\&theta;}}_k}{\mathrm{\&lambda;}}\right),$ Then have:

A(θ)＝[a(θ ₁)，a(θ ₂)，…，a(θ _K)] (4)

${a\left({\mathrm{\&theta;}}_k\right)}^T=\&lsqb;1,{\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1,...,v_l^l,...,v_{L-1}^{L-1}\&rsqb;---\left(5\right)$

A (θ _k) be reach wave angle be θ k reach the ripple direction vector.

s(n) ^T＝[s ₁(n)…s _K(n)] (6)

s _k(n) be that to reach wave angle be θ _kArrived wave signal base band complex envelope.

N(n) ^T＝[n ₁(n)n ₂(n)…n _L(n)] (7)

N (n) ^TIt is array antenna additive noise vector.

Under the effect of control section (4), switching matrix (2) is selected L from the signal that input 2-i1 to 2-iL receives _rIndividual, by output 2-1 to 2-L _rOutput.Node K in the switching matrix (2) ₁₁To K _LrLOperating state determine by the output signal 4-1 of control section (4).Control section (4) is by port 4-1 output control signal, and in return the node of matrix (2) is selected signal, is input to the control input end 2-c1 of switching matrix (2).Node K when switching matrix (2) _MnWhen not selected, K _Mn=0, the input 2-in of switching matrix (2) and output 2-m disconnect, input signal y _n(n) can not be sent to the output of switching matrix (2), this snap is rejected; Choose the node K of switching matrix (2) when control section (4) _MnThe time, node K _Mn=1, the input 2-in of switching matrix (2) is communicated with output 2-m, input signal y _n(n) the output port 2-m by switching matrix (2) exports.

In foundation array signal processing system of the present invention, according to array signal processing method disclosed by the invention, in each snap sampling constantly, control section (4) will be controlled the operating state of each node in the switching matrix (2), make whole output 2-1 to 2-L of switching matrix (2) _rAlways can with the L among switching matrix (2) the input 2-i1 to 2-iL _rIndividual input is communicated with, with remaining L-L _rIndividual input disconnects and connecting.The L of switching matrix (2) _rThe L of individual input and array antenna _rIndividual array element is corresponding one by one.Therefore, under the effect of control section (4), switching matrix (2) is L in the sampling of each snap constantly all can be array antenna _rOutput signal on the individual array element is sent to output 2-1 to 2-L undampedly _r

According to array signal processing method disclosed by the invention, in each snap sampling constantly, control section (4) is selected L from L input arrived wave signal of switching matrix (2) _rIndividual arrived wave signal sends signal receiver (3) to by switching matrix (2).

Switching matrix (2) is by output 2-1 to 2-L _rWith L _rThe arrived wave signal of individual array element is sent to the input 3-i1 to 3-iL of signal receiver part (3) respectively _r

In foundation array signal processing system of the present invention, signal receiver (3) comprises automatic gain control (AGC), demodulating unit, decoding unit, analog to digital conversion (A/D) circuit.Agc circuit carries out AGC to received signal to be handled, and makes the arrived wave signal amplitude of reception satisfy the job requirement of subsequent conditioning circuit; Demodulating unit carries out demodulation to the signal of handling through AGC, becomes baseband complex signal, and baseband complex signal contains the relative amplitude and the phase information of arrived wave signal; The A/D translation circuit is transformed to the digital baseband complex signal with the Analog Baseband complex signal, so that system applies DSP technology.Signal receiver (3) is by output 3-01 to 3-0L _rOutput arrived wave signal snap y ₁(n) to y _Lr(n).Wherein, y _l(n), l=1 ..., L _rCorresponding to any L in L the arrived wave signal _rIndividual arrived wave signal.

The output signal of signal receiver (3) is the digital baseband complex signal, by input 5-1 to 5-L _rBe input to covariance calculating section (5).Covariance calculating section (5) is finished auto-correlation and the cross-correlation calculation to input signal, and the matrix that its assessment of covariation that is output as each input signal is formed is defined as covariance matrix, R _YYThe form of covariance matrix:

In the formula, r _IjBe bay i, the covariance between the j dateout, y represents array antenna output signal vector, i, j ∈ (1 ..., Lr), Lr is an array number.

By formula (8) as seen, covariance matrix R _YYBe the covariance matrix of array antenna output signal vector on spatial domain, form by the covariance between each delayed output signals of antenna.

In foundation array signal processing system of the present invention, according to array signal processing method disclosed by the invention, constantly the data on the whole Lr array elements in the subarray are taken a sample simultaneously in each snap sampling, obtain an output snap of Lr array element, according to the error precision requirement, each sub antenna array is repeated the sampling of N correlations, obtain N snap of Lr array element, can calculate according to formula (8).Introduce concrete processing method below.

The covariance matrix R of array antenna output signal _YYBe defined as:

R _YY＝E{[A(θ)S(n)+N(n)][A(θ)S(n)+N(n)] ^H} (9)

Wherein E{} represents to get the statistical expectation of bracket interior element.Formula (9) can further be written as:

R _YY＝E{[A(θ)S(n)+N(n)][S(n) ^HA(θ) ^H+N(n) ^H]} ＝E{A(θ)S(n)S(n) ^HA(θ) ^H+N(n)S(n) ^HA(θ) ^H+A(θ)S(n)N(n) ^H+N(n)N(n) ^H} (10)

Under the separate condition of arrived wave signal and noise, the cross-product term in the formula (10) equals zero, so have

R _YY＝E{A(θ)S(n)S(n) ^HA(θ) ^H}+E{N(n)N(n) ^H} ＝A(θ)R _s(n)A(θ) ^H+σ ²I (11)

Wherein, I is Lr * Lr rank unit matrix.

By formula (11) as can be known, need ask source signal covariance matrix R earlier _s, just can obtain the covariance matrix R of array antenna output signal again according to formula (11) _YY

R _s＝E{[s ₁(n)…s _K(n)] ^T[s ₁(n)…s _K(n)] ^*} (12)

When disregarding noise, array antenna output covariance matrix is R _Xx:

R _xx＝A(θ)R _s(l)A(θ) ^H (13)

When considering additive noise, the covariance matrix of array antenna output signal is following form:

R _YY＝R _xx+σ ²I (14)

σ wherein ²Be noise variance, I is a unit matrix.

Can obtain the valuation of covariance matrix with N snap of array output at the mean value of time domain

${\hat{R}}_{\mathrm{YY}}=\frac{1}{N}\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}Y\left(n\right)Y{\left(n\right)}^H---\left(15\right)$

Be covariance matrix R _YYConsistent Estimation.

Can obtain the covariance matrix valuation of any one subarray antenna output according to formula (15).

Can determine required whole subarrays according to subarray antenna selection criterion disclosed by the invention and method.

According to signal processing method disclosed by the invention,, successively whole subarray antenna output signals are taken a sample according to the sequential of determining.Behind the snap sampling number of times through regulation, can obtain the output snap on whole subarray antennas, the whole array element.

According to signal processing method disclosed by the invention, under the coordination of control section (4), covariance calculating section (5) is obtained the covariance matrix valuation of each subarray antenna output snap one by one according to formula (15).Result of calculation is sent to data storage part (7) by the output 5-0 output of covariance matrix calculating section (5), and as the intermediate object program of calculating, each subarray output covariance matrix valuation is temporary among the data storage (7).

The output of each sub antenna array produces a covariance matrix.Generally speaking, the quantity of subarray antenna equals $C=C_L^{\mathrm{Lr}}.$ Wherein, L is main array antenna array number, and Lr is a subarray bay number, and it equals signal receiver quantity, C _L ^LrrIt is the number of combinations of selecting Lr element in L the element.

Signal processing method disclosed by the invention is: the covariance matrix that calculates each subarray array element output; According to signal processing criterion disclosed by the invention,, construct main array antenna output equivalent covariance matrix by subarray output covariance matrix based on the subarray synthetic technology; Use based on the algorithm of eigenmatrix singular value decomposition arrived wave signal space characteristics parameter is estimated, realize reaching ripple DOA and estimate.The detailed construction method of main array antenna equivalence covariance matrix is described with an embodiment below.

With reference to Fig. 1, to establish main array antenna array number and equal L=4, the subarray bay is counted Lr=2.Sequence number is 1 array element array element for referencial use, and the array element number consecutively is 1,2,3,4.Then the subarray antenna number equals C ₄ ²=6, six subarrays are respectively by array element (1,2), and (2,3), (3,4) (4,1), (1,3), (2,4) are formed, and subarray stream shape is respectively:

$A_1\left(\mathrm{\&theta;}\right)=\&lsqb;a_1\left({\mathrm{\&theta;}}_1\right),a_1\left({\mathrm{\&theta;}}_2\right),a_1\left({\mathrm{\&theta;}}_3\right)\&rsqb;=\left[\begin{array}{ccc}1& 1& 1\\ v_1& v_2& v_3\end{array}\right]---\left(16\right)$

$A_2\left(\mathrm{\&theta;}\right)=\&lsqb;a_2\left({\mathrm{\&theta;}}_1\right),a_2\left({\mathrm{\&theta;}}_2\right),a_2,\left({\mathrm{\&theta;}}_3\right)\&rsqb;=\left[\begin{array}{ccc}{v}_1& v_2& v_3\\ {\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^2& {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_2^2& {\mathrm{<figure-callout\; id="3"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_3^2\end{array}\right]---\left(17\right)$

$A_3\left(\mathrm{\&theta;}\right)=\&lsqb;a_3\left({\mathrm{\&theta;}}_1\right),a_3\left({\mathrm{\&theta;}}_2\right),a_3\left({\mathrm{\&theta;}}_3\right)\&rsqb;=\left[\begin{array}{ccc}{\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^2& {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_2^2& {\mathrm{<figure-callout\; id="3"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_3^2\\ {\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^3& {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_2^3& {\mathrm{<figure-callout\; id="3"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_3^3\end{array}\right]---\left(18\right)$

$A_4\left(\mathrm{\&theta;}\right)=\&lsqb;a_4\left({\mathrm{\&theta;}}_1\right),a_4\left({\mathrm{\&theta;}}_2\right),a_4\left({\mathrm{\&theta;}}_3\right)\&rsqb;=\left[\begin{array}{ccc}{\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^3& {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_2^3& {\mathrm{<figure-callout\; id="3"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_3^3\\ 1& 1& 1\end{array}\right]---\left(19\right)$

$A_5\left(\mathrm{\&theta;}\right)=\&lsqb;a_5\left({\mathrm{\&theta;}}_1\right),a_5\left({\mathrm{\&theta;}}_2\right),a_5\left({\mathrm{\&theta;}}_3\right)\&rsqb;=\left[\begin{array}{ccc}1& 1& 1\\ {\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^2& {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_2^2& {\mathrm{<figure-callout\; id="3"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_3^2\end{array}\right]---\left(20\right)$

$A_6\left(\mathrm{\&theta;}\right)=\&lsqb;a_6\left({\mathrm{\&theta;}}_1\right),a_6\left({\mathrm{\&theta;}}_2\right),a_6\left({\mathrm{\&theta;}}_3\right)\&rsqb;=\left[\begin{array}{ccc}{v}_1& v_2& v_3\\ {\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^3& {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_2^3& {\mathrm{<figure-callout\; id="3"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_3^3\end{array}\right]---\left(21\right)$

The covariance matrix of subarray antenna output is R _YYi:

R _YYi＝A _i(θ)R _SiA _i(θ) ^H+σ ²I (22)

Wherein I is 2 * 2 rank unit matrix, i=1, and 2 ..., 6.

In form, R _SiWith R _sBeing identical, all is K * kth moment battle array, is 3 * 3 rank here, is the cross-correlation matrix between the source signal, and it is mutual when uncorrelated to reach ripple in information source, R _Si, R _sBe the diagonal angle square formation.In fact the corresponding constantly arrived wave signal S of different snap samplings _i(n) and additive noise N _i(n) incomplete same, so R _Si≠ R _s, i represents i subarray.Construct equivalent main array in order to use based on the subarray synthetic method, require ripple reach signal be broad sense steadily, in the random process of spatial domain and time domain traversal.At this moment, can think R _Si=R _s, σ _i ²=σ ², σ wherein _i ²Be the variance of subarray antenna i additive noise, σ ²It is the variance of main array antenna additive noise.

In arrived wave signal processing method disclosed by the invention, the sampling method and the process of regulation subarray output snap are as follows: control section (4) sends a control command to switching matrix (2) constantly in each sampling, choose a wherein group node, Lr output node of switching matrix (2) is communicated with Lr input node, the output snap on Lr the array element is sent to Lr signal receiver (3).

Control section (4) is chosen Lr the input node that links to each other with first subarray bay in the switching matrix (2) constantly in first snap sampling, constantly choose Lr the input node that links to each other with second sub-array antenna array element in the switching matrix (2) second snap sampling, ..., and the like, circulation is carried out.

Like this, what signal receiver (3) constantly received signal in the sampling of first snap is the output snap of first subarray antenna, and what constantly receive signal second snap sampling is the output snap of second sub-array antenna.。。, what constantly receive signal in the sampling of n snap is the output snap of n sub-array antenna.

Signal receiver (3) continuous operation is taken a sample to the whole Lr of first subarray array element outputs constantly in first snap sampling; Second snap sampling taken a sample to second whole Lr of subarray array element outputs constantly ..., n snap sampling is constantly to n the whole Lr of subarray array element output samplings.So successively Lr array element dateout of each subarray antenna sampled.Finish after the sampling process to Lr array element output of last subarray antenna, restart Lr array element output sampling to first subarray antenna again, so circulation is carried out.Can require to determine hits according to different valuation error precisions.Each sampling sampling snap constantly is temporary in the data storage (7).

In arrived wave signal processing method disclosed by the invention, definition: whole subarray antennas required fast umber of beats of once sampling respectively is called sampling period of subarray antenna.Therefore, equal based on the C of the direct computing method of main array output doubly based on the synthetic sampling period of subarray, wherein C is the subarray antenna amount.

Based on the output of signal receiver (3), covariance calculating section (5) can go out to calculate the output signal covariance matrix of subarray antenna in proper order, is respectively R _YY1, R _YY2, R _YY3, R _YY4, R _YY5, R _YY6:

${\mathrm{<figure-callout\; id="1"\; label="R\; YY"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{YY}}=\left[\begin{array}{ccc}1& 1& 1\\ v_1& v_2& v_3\end{array}\right]E\left\{\left[\begin{array}{ccc}{S}_1{\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_1^{*}& S_1{\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_2^{*}& S_1{\mathrm{<figure-callout\; id="3"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_3^{*}\\ S_2{\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_1^{*}& S_2{\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_2^{*}& S_2{\mathrm{<figure-callout\; id="3"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_3^{*}\\ S_3{\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_1^{*}& S_3{\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_2^{*}& S_3{\mathrm{<figure-callout\; id="3"\; label="S"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">S</figure-callout>}}_3^{*}\end{array}\right]\right\}\left[\begin{array}{cc}1& {\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^{*}\\ 1& {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_2^{*}\\ 1& {\mathrm{<figure-callout\; id="3"\; label="v"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">v</figure-callout>}}_3^{*}\end{array}\right]+{\mathrm{\&sigma;}}^{2}I=\left[\begin{array}{cc}{r}_{11}& r_{12}\\ r_{21}& r_{22}\end{array}\right]---\left(23\right)$

Unification is write as following form:

${\mathrm{<figure-callout\; id="1"\; label="R\; YY"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{YY}}=\left[\begin{array}{cc}{r}_{11}\left(1\right)& r_{12}\left(1\right)\\ r_{21}\left(1\right)& r_{22}\left(1\right)\end{array}\right]---\left(24\right),$

${\mathrm{<figure-callout\; id="2"\; label="R\; YY"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{YY}}=\left[\begin{array}{cc}{r}_{22}\left(2\right)& r_{23}\left(1\right)\\ r_{32}\left(1\right)& r_{33}\left(1\right)\end{array}\right]---\left(25\right),$

${\mathrm{<figure-callout\; id="3"\; label="R\; YY"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{YY}}=\left[\begin{array}{cc}{r}_{33}\left(2\right)& r_{34}\left(1\right)\\ r_{43}\left(1\right)& r_{44}\left(1\right)\end{array}\right]---\left(26\right),$

${\mathrm{<figure-callout\; id="4"\; label="R\; YY"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{YY}}=\left[\begin{array}{cc}{r}_{44}\left(2\right)& r_{41}\left(1\right)\\ r_{14}\left(1\right)& r_{11}\left(2\right)\end{array}\right]---\left(27\right),$

${\mathrm{<figure-callout\; id="5"\; label="R\; YY"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{YY}}=\left[\begin{array}{cc}{r}_{11}\left(3\right)& r_{13}\left(1\right)\\ r_{31}\left(1\right)& r_{33}\left(3\right)\end{array}\right]---\left(28\right),$

${\mathrm{<figure-callout\; id="6"\; label="R\; YY"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">R</figure-callout>}}_{\mathrm{YY}}=\left[\begin{array}{cc}{r}_{22}\left(3\right)& r_{24}\left(1\right)\\ r_{42}\left(1\right)& r_{44}\left(3\right)\end{array}\right]---\left(29\right)$

In the formula, r _Ij(k) be covariance between array element i and the array element j output signal, k represents r _Ii(k) this number of times that in a sampling period of subarray, occurs, for example r ₁₁(2) be illustrated in the sampling period and occur covariance item r for the second time ₁₁In the present embodiment, a sampling period of subarray equals 6 snap sampling periods, supposes that sampling frequency equals 1.

Covariance calculating section (5) passes through data/address bus 5-0 with result of calculation R _YY1, R _YY2, R _YY3, R _YY4, R _YY5, R _YY6Send to the input/output port 7-1 of data storage (7).

Control section (4) is read subarray covariance matrix R from data storage (7) _YY1, R _YY2, R _YY3, R _YY4, R _YY5, R _YY6, data are sent to the port 6-1 of subarray composition part (6) by bus 7-2.

In foundation signal processing system of the present invention, subarray composition part (6) is based on the covariance matrix R of subarray _YY1, R _YY2, R _YY3, R _YY4, R _YY5And R _YY6, according to signal processing method disclosed by the invention, the synthetic covariance matrix valuation R that is equivalent to main array output _YY

$R_{\mathrm{YY}}=\left[\begin{array}{cccc}{r}_{11}\left(3\right)& r_{12}& r_{13}& r_{14}\\ r_{21}& r_{22}\left(3\right)& r_{23}& r_{24}\\ r_{31}& r_{32}& r_{33}\left(3\right)& r_{34}\\ r_{41}& r_{42}& r_{43}& r_{44}\left(3\right)\end{array}\right]+{\mathrm{\&sigma;}}^{2}I---\left(30\right)$

In the formula, r _Ij(n) n in represents r _IjThe number of times that this repeats in the subarray covariance is calculated.According to the control command that control section (4) sends to subarray composition part (6) by port 4-3, subarray composition part (6) is with equivalent covariance matrix R _YYBy port 6-1 output, the data/address bus 7-2 input by memory portion (7) is temporary in the memory (7).

In according to signal processing system of the present invention and signal processing method, the covariance item that repeats is taked following processing method:

1) keep the value that occurs for the first time, the value of abandoning repeating:

r _ij＝r _ij(1) (31)

r _Ij(1) is the value that occurs first.

2) or, get the value of the arithmetic mean of each time valuation as this:

$r_{\mathrm{ij}}=\frac{1}{N}{\mathrm{\&Sigma;}}_{n=1}^N{r}_{\mathrm{ij}}\left(n\right)---\left(32\right)$

N is r _Ij(n) number of times of Chu Xianing.

3) or, the mean value of N valuation of getting equivalent covariance matrix is as covariance matrix:

$R_{\mathrm{YY}}=\frac{1}{N}{\mathrm{\&Sigma;}}_{l=1}^N{R}_{\mathrm{YY}}\left(l\right)---\left(33\right)$

N is the sampling period number.

When calculating main array equivalence covariance matrix based on the subarray synthetic method, need C fast beat of data, wherein, first fast beat of data is the sampling value of first subarray antenna output, second fast beat of data is the sampling value of second sub-array antenna output, ..., C fast beat of data is the sampling value of C sub-array antenna output.

Compare with other signal processing system, by increasing the fast umber of beats of sampling, proposed by the invention has following advantage based on subarray synthetic signal processing system and signal processing method: make the quantity of desired signal receiver (3) be reduced to Lr from L, Lr＜L, simultaneously, do not reduce the main performance of signal processing system, as resolution, beamwidth etc.

Subarray composition part (6) outputs to data storage (7) afterwards with equivalent covariance matrix data, and control section (4) sends instruction by port 4-4 to data storage (7), and 7-c1 is input to data storage by port.This instruction makes data storage (7) from port 7-3 equivalent covariance matrix be exported, and is sent to space characteristics test section (8) by port 8-2.

Space characteristics test section (8) is by the instruction of port 8-1 reception from control section (4), read in based on the synthetic equivalent covariance matrix data of subarray from port 8-2, according to DOA method of estimation based on the eigenmatrix singular value decomposition, equivalent covariance matrix is carried out singular value decomposition, extract the matrix exgenvalue valuation, carry out parameter Estimation, realize the estimation of arrived wave signal DOA super-resolution.Ripple reaches the port 8-3 output of angular data by space characteristics test section (8).

In according to signal processing system of the present invention and signal processing method, reach ripple DOA estimation approach and be not limited to the MUSIC algorithm that the embodiment of the invention adopts.

Signal processing system disclosed in this invention and signal processing method are applicable to any method for parameter estimation based on Singular Value Decomposition Using, estimate wave beam formation etc. comprising reaching ripple DOA.

Optimize the scheme number of choosing the subarray antenna:

In the embodiment of above-mentioned foundation signal processing system of the present invention and signal processing method, the subarray antenna amount is C, $C=C_L^{\mathrm{Lr}}---\left(34\right).$

Discover the array number Lr that comprises when subarray〉2 the time, when calculating equivalent covariance matrix, required subarray antenna amount can be less than C _L ^LrIn the content below, the present invention will disclose the computational methods formula of required subarray antenna minimum number; Also will openly choose the criterion of required subarray antenna, and be called to optimize and choose criterion.

Optimize and choose criterion:

1. the number of times that occurs in different subarray antennas of any one array element is few as far as possible.

2. any two array elements occur once in a sub-array antenna at least simultaneously.

3. count under the selectable situation of Lr at the subarray bay, get the selectable maximum of Lr.

4. under the condition that main array array number L and subarray array number Lr determine, get the minimum scheme of choosing of subarray antenna number.

Choose criterion according to this optimization, can optimize the scheme of choosing and count K and be:

$K=C_{L-\mathrm{<figure-callout\; id="2"\; label="Lr\; +"\; filenames="C20041000967900201.png"\; state="\{\{state\}\}">Lr</figure-callout>}+2}^2---\left(35\right)$

For reducing the disturbance of random noise, can limit minimum number P that array element occurs greater than certain threshold value in subarray, for example P 〉=50, the i.e. minimum number of coherent sampling to covariance.

Embodiment performance simulation result:

Accompanying drawing 2-5 compares with the ripple DOA estimation performance simulation result that reaches that adopts main array output based on the synthetic ripple DOA estimated result that reaches of subarray for employing.

Adopt the MUSIC algorithm, to carrying out emulation, pseudo-spectral function P based on subarray synthetic arrived wave signal treatment system and signal processing method according to of the present invention _MU(θ) be

$P_{\mathrm{MU}}\left(\mathrm{\&theta;}\right)=\frac{1}{\underset{k=K+1}{\overset{M}{\mathrm{\&Sigma;}}}{\left|e_k^{H}a\left(\mathrm{\&theta;}\right)\right|}^2}---\left(36\right)$

Simulated conditions: 3 information sources reach ripple, and main array antenna is made up of four array elements, and the subarray antenna comprises two array elements.

Each picture group comprises two sets of curves respectively.Wherein, last figure is based on the synthetic DOA estimated result of subarray, and figure below is the DOA estimated result based on main array output.

Fig. 2: it is 15,45 ,-45 degree that fast umber of beats N=40, signal to noise ratio snr=20 dB, ripple reach the angle.

Fig. 3: it is 15,19 ,-60 degree that fast umber of beats N=400, signal to noise ratio snr=20 dB, ripple reach the angle.

Fig. 4: it is 60,64 ,-30 degree that fast umber of beats N=50, signal to noise ratio snr=100 dB, ripple reach the angle.

Fig. 5: it is 60,64 ,-30 degree that fast umber of beats N=10000, signal to noise ratio snr=30 dB, ripple reach the angle.

The embodiment simulation result shows:

Among Fig. 2, the DOA angle of arrived wave signal differs bigger.The result shows, under medium signal to noise ratio condition, use less sampling snap just can estimate exactly that efferent echo reaches the angle, from two figure up and down as can be known, the DOA estimated result of exporting fast beat of data based on synthetic DOA estimated result of subarray and the direct main array of employing is identical.

Among Fig. 3, the DOA angle of arrived wave signal differs smaller.The result shows, can tell different arrived wave signals based on the DOA algorithm for estimating that subarray is synthetic.From two figure up and down as can be known, under identical signal to noise ratio condition, needing increases fast umber of beats, just can make the DOA estimated result more accurate.

Among Fig. 4, it is identical with Fig. 3 that signal reaches wave angle, also gets smaller value.Compare with Fig. 3, simulated conditions changes into: reduce the fast umber of beats of sampling, increase the arrived wave signal signal to noise ratio.The result shows, under the high s/n ratio condition, what very little fast umber of beats just can be told close proximity reaches the ripple direction.

Among Fig. 5, it is bigger that signal reaches wave angle, and adjacent to reach wave angle degree difference less.Signal to noise ratio is got representative value, increases fast umber of beats.The result shows, and is effective equally based on the DOA algorithm for estimating that subarray is synthetic.

Comparison diagram 5 and Fig. 3 as can be known, it is influential to DOA resolution that signal reaches wave angle.Reach wave angle and increase, DOA resolution will reduce.

Fig. 2 to Fig. 5 shows that disclosed in this invention is effective based on subarray synthetic arrived wave signal treatment system and signal processing method.

Although disclose most preferred embodiment of the present invention and accompanying drawing for the purpose of illustration, it will be appreciated by those skilled in the art that: without departing from the spirit and scope of the invention and the appended claims, various replacements, variation and modification all are possible.Therefore, the present invention should not be limited to most preferred embodiment and the disclosed content of accompanying drawing.

Claims (9)

1. an array antenna signal processing system based on the synthetic arrived wave signal direction determining method of subarray, realizes that receiver quantity is less than the arrived wave signal direction estimation under the bay quantity situation, and this treatment system comprises:

The array antenna part, the bay induction receives the far field arrived wave signal;

The switching matrix part is according to the instruction connection or the disconnected node of control circuit;

The signal receiver part is amplified or decay bay amplitude output signal, carries out the automatic gain control and treatment, carries out demodulation, decoding is transformed to digital baseband signal with arrived wave signal;

Control section, the connection or the disconnection of control switching matrix, the accessing operation of control data memory;

The covariance calculating section, the covariance of calculating delayed output signals;

Subarray composition part, synthetic equivalent covariance;

The data storage part, storing intermediate result of data operation and final result;

The space characteristics test section, the space characteristic parameter of detection arrived wave signal is realized the estimation of arrived wave signal direction.

2. array antenna signal processing system according to claim 1 is characterized in that: the array antenna type is selected from uniform line-array, perhaps the array of plane distribution.

3. array antenna signal processing system according to claim 1 is characterized in that: array antenna is minimum to comprise 2 array elements.

4. array antenna signal processing system according to claim 1 is characterized in that: receiver quantity equals bay quantity.

5. array antenna signal processing system according to claim 1 is characterized in that: the array element of subarray antenna adopts time division multiplexing mode to be connected with the input of receiver by switching matrix.

6. realize reaching ripple direction super-resolution estimated signals processing method based on the arrived wave signal snap for one kind, described method comprises:

According to the array element numbering, main array antenna is divided into a plurality of subarray antennas;

Based on the time division multiplexing processing method, ask the covariance matrix of sub antenna array output;

Ask the equivalent covariance matrix of main antenna array output;

The arrived wave signal direction determining method that decomposes based on proper subspace carries out the estimation of arrived wave signal direction.

7. according to claim 6 realization reaches ripple direction super-resolution estimated signals processing method based on the arrived wave signal snap, it is characterized in that: receiver is gathered the data of the whole array elements outputs of sub-array antenna constantly in each snap sampling.

It is 8. according to claim 6 that realization reaches ripple direction super-resolution estimated signals processing method based on the arrived wave signal snap, it is characterized in that: pass through time-sharing multiplex, receive the fast beat of data of output of handling any amount bay, realize reception and processing a large amount of delayed output signals.

9. the method for sub-array antenna is chosen in an optimization, main array antenna is divided into a plurality of subarray antennas, the array number of each subarray antenna is less than the array number of main array antenna, the array number of each subarray antenna equals the quantity of signal receiver, the array element of subarray antenna is passed through switching matrix, adopt time division multiplexing mode to be connected, it is characterized in that, adopt following principle to divide with the input of receiver:

Export the array antenna output correlation matrix of the synthetic and main array output of fast beat of data correlation matrix equivalence by whole subarrays, be called equivalent main array output correlation matrix;

Any two array elements of main array antenna occur once in some subarray antennas at least simultaneously; The quantity K of subarray antenna satisfies following formula: $K=C_{L-\mathrm{Lr}+2}^2$

Wherein, C represents combination calculation, and L represents the quantity of the array element of main array antenna, and Lr represents the quantity of the array element of certain subarray antenna.

Images