CN104199020A - Multi-frame information fusion based meter wave array radar target elevation measuring method - Google Patents

Abstract

The invention belongs to the technical field of radar target elevation measurement and particularly relates to a multi-frame information fusion based meter wave array radar target elevation measuring method. The multi-frame information fusion based meter wave array radar target elevation measuring method comprises the following steps of obtaining the beam width of a radar beam scanning chart and the number of peak points in the radar beam scanning chart; determining the number of angle search intervals contained within an airspace action angle range at the elevation of a radar and the number of search intervals contained in the beam width of the radar beam scanning chart; obtaining echo data of distance units on which K frames of targets are arranged; obtaining a beam scanning matrix; performing beam forming on the echo data of the distance units on which the K frames of targets are arranged through the beam scanning matrix, obtaining a beam forming vector of every frame of data and forming into a beam forming matrix; extracting serial numbers of m data with maximum amplitude in every beam forming vector and forming into an information matrix; performing information fusion on the information matrix and obtaining a target elevation measured value.

Description

Metric wave array radar target elevation measuring method based on multiframe information fusion

Technical field

The invention belongs to radar target measurement of elevation technical field, particularly the metric wave array radar target elevation measuring method based on multiframe information fusion.

Background technology

Metre wave radar has unique advantage at the aspect such as over-the-horizon detection, anti-electronic interferences, is generally paid attention in recent years.But because the radar emission signal wavelength of metric wave system is longer, according to antenna theory, the physical pore size of antenna beamwidth and antenna is directly proportional, and under identical antenna aperture, the beam angle of metric wave radar increases greatly with respect to the beam angle of microwave radar.Therefore, in the time surveying low elevation angle target, radar beam is beaten ground, the multipath reflection ripple of the direct wave of target and ground (sea) face reflection superposes in antenna beam main lobe, as shown in Figure 1, under some incident elevation angle, direct-path signal and multipath signal are cancelled out each other, and the elevation angle (highly) measurement to target is had difficulties.

For this problem, J.Litva is at document " Use of a highly deterministic multipath signal model in low angle tracking ", IEEE Radar Signal Processing, Vol.138, pp.163-171, a kind of accurately maximum likelihood (RML is proposed Apr.1991., Refined maximum likelihood algorithm) method, the method is utilized the relation between direct-path signal and multipath signal, the associating steering vector of structure direct-path signal multipath signal, then utilize the associating steering vector of structure to carry out beam scanning to receiving data, thereby draw angle on target.In the time that the signal to noise ratio (S/N ratio) of radar return signal is higher, RML method has good angle measurement accuracy, but, while adopting RML method to process to received signal, the beam scanning figure that receives signal there will be the graing lobe that amplitude is very high, in low signal-to-noise ratio situation, these graing lobes can have a strong impact on the measurement of radar to target elevation (highly).

Summary of the invention

The object of the invention is to for above-mentioned the deficiencies in the prior art, propose the metric wave array radar target elevation measuring method based on multiframe information fusion, to improve the measurement of angle performance of traditional RML method.

For realizing above-mentioned technical purpose, the present invention adopts following technical scheme to be achieved.

Realizing technical scheme of the present invention is: by information fusion, eliminate the impact that in RML method, wave beam raster lobe is measured angle on target, realize the high-acruracy survey to target elevation, its step comprises as follows:

Step 1, draws the peak point number m in beam angle B and the radar beam scintigram of radar beam scintigram;

Step 2, determines the scouting interval comprising in the beam angle B of angle searching space-number L that radar comprises in the spatial domain of pitching dimension effect angular range and radar beam scintigram and counts M;

Step 3, utilizes radar receiving target echo data, obtains the echo data of K frame target place range unit, and wherein, the echo data of k frame target place range unit is expressed as x _k, k=1,2 ..., K;

Step 4, according to the array number N of radar emission signal wavelength lambda, radar array antenna, known multipath reflection coefficient ρ, radar acts on angular range lower bound θ in the spatial domain of pitching dimension _d, radar is at angle searching precision Δ θ and the high H of radar antenna frame of pitching dimension, draws beam scanning matrix W;

Step 5, utilizes beam scanning matrix W respectively the echo data of K frame target place range unit to be carried out to wave beam formation, and the wave beam that obtains each frame data forms vector, and wherein, it is z that the wave beam of the echo data of k frame target place range unit forms vector _k, by z ₁to z _kform beam forming matrix Z;

Step 6, extracts each wave beam and forms the sequence number of m data of amplitude maximum in vector, utilizes the sequence number extracting to form corresponding frame to receive the maximum value vector of data, and the maximum value vector of k frame reception data is v _k; By v ₁to v _kconfiguration information matrix V;

Step 7, carries out information fusion to information matrix V, obtains target elevation measured value

Beneficial effect of the present invention is: 1) the present invention is according to the correlativity of target information in adjacent multiframe radar return data, the target information that adjacent multiframe is received in data merges, reduce the impact that in RML method, beam scanning secondary lobe is measured angle on target, improved angle measurement accuracy.2) the present invention, by the fusion to multiframe information, has improved the utilization factor of target information, in the time that echoed signal is weak, still has good measurement of angle performance.

Brief description of the drawings

Fig. 1 is that multipath signal is propagated schematic diagram;

Fig. 2 is the process flow diagram of the metric wave array radar target elevation measuring method based on multiframe information fusion of the present invention;

Fig. 3 is for adopting the present invention and traditional RML method angle on target to be carried out to the measurement result schematic diagram of the target elevation that 100 secondary trackings measure in emulation 1;

Fig. 4 adopts the present invention and traditional RML method to carry out angle measurement root-mean-square error that 1000 independent measurements the draw variation relation schematic diagram with signal to noise ratio (S/N ratio) to target azimuth angle in emulation 2.

Embodiment

Below in conjunction with accompanying drawing, the invention will be further described:

With reference to Fig. 2, it is the process flow diagram of the metric wave array radar target elevation measuring method based on multiframe information fusion of the present invention.Should the metric wave array radar target elevation measuring method based on multiframe information fusion comprise the following steps:

Step 1, draws the peak point number m in beam angle B and the radar beam scintigram of radar beam scintigram.

Its concrete sub-step is:

1a) radar emission signal wavelength is λ, and radar array antenna is the even linear array being made up of N array element, and array element distance is λ/2, and radar antenna frame height is H, draws the beam angle B of radar beam scintigram, B=50/ (H/ λ+N).

Be 1b) [θ by radar at the spatial domain of pitching dimension effect angular range _d, θ _u], at the spatial domain of pitching dimension effect angular range, obtain the peak point number m in radar beam scintigram by radar, wherein expression rounds up.

Step 2, determines the scouting interval comprising in the beam angle B of angle searching space-number L that radar comprises in the spatial domain of pitching dimension effect angular range and radar beam scintigram and counts M.

Its concrete sub-step is:

Angle searching precision Δ θ 2a) being tieed up by radar pitching, obtains radar at the spatial domain of pitching dimension effect angular range [θ _d, θ _u] in the angle searching space-number L that comprises,

2b) the angle searching precision Δ θ in pitching dimension by radar, obtains the angle searching space-number M that beam angle B comprises,

Step 3, utilizes radar receiving target echo data, obtains the echo data of K frame target place range unit, and wherein, the echo data of k frame target place range unit is expressed as x _k, wherein, k=1,2 ..., K.

Step 4, according to the array number N of radar emission signal wavelength lambda, radar array antenna, known multipath reflection coefficient ρ, radar acts on angular range lower bound θ in the spatial domain of pitching dimension _d, radar is at angle searching precision Δ θ and the high H of radar antenna frame of pitching dimension, draws beam scanning matrix W.

Its concrete sub-step is:

4a) initialization scouting interval counter, represents scouting interval count value with γ, γ=1, and 2 ..., in the time of γ=1, carry out sub-step 4b).

4b) in the bay of radar array antenna, an optional array element is as with reference to array element; By radar emission signal wavelength lambda and bay number N, structure weight vector w _{γ 1}with weight vector w _{γ 2}, wherein,

$w_{\mathrm{\γ}1}=[e^{j2\mathrm{\π}\frac{d_1}{\mathrm{\λ}}\left[\sin ((\mathrm{\γ}-1)\mathrm{\Δ\θ}-{\mathrm{\θ}}_d)\right]},...,e^{j2\mathrm{\π}\frac{d_n}{\mathrm{\λ}}\left[\sin ((\mathrm{\γ}-1)\mathrm{\Δ\θ}-{\mathrm{\θ}}_d)\right]},...,e^{j2\mathrm{\π}\frac{d_N}{\mathrm{\λ}}\left[\sin ((\mathrm{\γ}-1)\mathrm{\Δ\θ}-{\mathrm{\θ}}_d)\right]}{]}^T$

$w_{\mathrm{\γ}2}=[e^{j2\mathrm{\π}\frac{d_1}{\mathrm{\λ}}\left[\sin ((\mathrm{\γ}-1)\mathrm{\Δ\θ}-{\mathrm{\θ}}_d)\right]},...,e^{j2\mathrm{\π}\frac{d_n}{\mathrm{\λ}}\left[\sin ((\mathrm{\γ}-1)\mathrm{\Δ\θ}-{\mathrm{\θ}}_d)\right]},...,e^{j2\mathrm{\π}\frac{d_N}{\mathrm{\λ}}\left[\sin ((\mathrm{\γ}-1)\mathrm{\Δ\θ}-{\mathrm{\θ}}_d)\right]}{]}^T$

Wherein, j is imaginary unit, d _nfor the distance between n array element and the reference array element of radar array antenna, n=1,2 ..., N; The transposition of subscript T representing matrix or vector.

4c), by the high H of radar array antenna holder, obtain the phase differential between direct-path signal and multipath signal for:

4d) by known multipath reflection coefficient ρ, weight vector w _{γ 1}with weight vector w _{γ 2}, obtain beam scanning weight vector w _γ, w _γfor:

4e) make the value of γ from increasing 1, and judge that whether scouting interval counter γ is greater than L, if so, obtains beam scanning matrix W, W=[w ₁, w ₂..., w _l], then perform step 5; If γ is not more than L, be back to sub-step 4b).

Step 5, utilizes beam scanning matrix W respectively the echo data of K frame target place range unit to be carried out to wave beam formation, and the wave beam that obtains each frame data forms vector, and wherein, it is z that the wave beam of the echo data of k frame target place range unit forms vector _k, by z ₁to z _kform beam forming matrix Z.

Its concrete sub-step is:

5a) initialization frame counter, represents frame count value with k, k=1, and 2 ..., in the time of k=1, carry out sub-step 5b).

5b) utilize the echo data x of beam scanning matrix W to k frame target place range unit _kcarry out beam scanning, obtain wave beam and form vector z _k, z _k=W ^hx _k, wherein, the conjugate transpose of subscript H representing matrix.

5c) make the value of k from increasing 1, and judge whether k is greater than K, if so, obtain beam forming matrix Z, Z=[z ₁, z ₂..., z _k], then perform step 6; If k is not more than K, be back to sub-step 5b).

Step 6, extracts each wave beam and forms the sequence number of m data of amplitude maximum in vector, utilizes the sequence number extracting to form corresponding frame to receive the maximum value vector of data, and the maximum value vector of k frame reception data is v _k; By v ₁to v _kconfiguration information matrix V.

Its concrete sub-step is:

6a) initialization frame counter, represents frame count value with k, k=1, and 2 ..., in the time of k=1, carry out sub-step 6b).

A 6b) initialization peak point counter, represents peak point number count value with η, η=1, and 2 ..., in the time of η=1, carry out sub-step 6c).

6c) find out wave beam and form vector z _kthe sequence number I of the data of middle amplitude maximum _η, wave beam forms vector z _kin I _ηindividual data are that wave beam forms vector z _kthe data of middle amplitude maximum.By I _ηreceive data maximum value vector v as k frame _kη component v _{η k}, wave beam is formed to vector z _kin with I _η2M+1 centered by individual data data value zero setting, wherein, M is the angle searching space-number that beam angle B comprises.

Specifically, wave beam is formed to vector z _kin with I _ηthe process of the 2M+1 centered by individual data data value zero setting is: wave beam is formed to vector z _kin I _ηindividual data zero setting, forms vector z by wave beam _kin I _η-M data to the I _η-1 data zero setting (is worked as I especially, _ηwhen <M, wave beam is formed to vector z _kin the 1st data to I _η-1 data zero setting), wave beam is formed to vector z _kin I _η+ 1 data to the I _η+ M data zero setting (is worked as I especially, _ηthe value of+M is greater than wave beam and forms vector z _kelement number D time, wave beam is formed to vector z _kin I _η+ 1 data to the D data zero setting).

6d) make the value of η from increasing 1, judge whether η is greater than m, if so, obtain k frame and receive data maximum value vector v _k, v _k=[v _1k, v _2k..., v _mk] ^t, then carry out sub-step 6e); If η is not more than m, be back to sub-step 6c).

The value that 6e) makes k is from increasing 1, and whether k be greater than K, if so, obtains information matrix V, V=[v ₁, v ₂..., v _k], then perform step 7; If k is not more than K, be back to sub-step 6b).

Step 7, carries out information fusion to information matrix V, obtains target elevation measured value

Information matrix V is carried out to information fusion multiple prior art, as weighted average method, the Bayes estimation technique and Markov chain method etc., adopts method of weighted mean to carry out information fusion to information matrix V in this method, and its concrete sub-step is:

7a) loop initialization counter l, represents loop count with l, l=1, and 2 ..., in the time of l=1, carry out sub-step 7b).

7b) the degrees of fusion matrix A of computing information matrix V, degrees of fusion matrix A is the matrix of K × K dimension, when natural number p gets the arbitrary numerical value in 1 to m, all has | v _p? _i-v _p? _k| when≤N, the capable k column element of the i a of degrees of fusion matrix A _ikbe 1; Otherwise the capable k column element of the i a of degrees of fusion matrix A _ikbe 0.Wherein, i=1,2 ..., K, k=1,2 ..., K, v _pi is the capable i column element of p of information matrix V, || represent to take absolute value.Be the capable k column element of the i a of degrees of fusion matrix A _ikfor:

Whether all elements that 7c) judges degrees of fusion matrix A all equals 1, if so, shows that each row data of information matrix V merge mutually, now obtains the angle index value S of target, wherein v _1ifor information matrix V the 1st row i column element, then carry out sub-step 7k); If all elements of degrees of fusion matrix A is not 1 entirely, carry out sub-step 7d).

7d) find out in degrees of fusion matrix A, the rower r of the row that nonzero element number is maximum,

$r=\underset{i}{\arg \max }\mathrm{sum}\left(a_i\right)$

Wherein, a _ifor the vector of the capable composition of i of degrees of fusion matrix A, sum () represent to ask each element in vector and;

Then, upgrade degrees of fusion matrix A, the process of upgrading degrees of fusion matrix A is: the vectorial a that judges the capable composition of r of degrees of fusion matrix A _rq component whether be 0, q=1,2 ..., K; If the vectorial a of the capable composition of the r of degrees of fusion matrix A _rq component be not 0, keep the q column data of information matrix V constant; If the vectorial a of the capable composition of the r of degrees of fusion matrix A _rq component be 0, the q of lastest imformation matrix V row, the process of the q row of lastest imformation matrix V is: the q column data of information matrix V is carried out to ring shift, that is: works as j=1,2 ..., when m-1, make v _j? _q=v _j+1q; If j=m, makes v _j? _q=v _1q; Wherein, v _j? _qrepresent the capable q column element of j of information matrix V.

7e) make the value of l from increasing 1, judge whether l is greater than m, if so, carry out sub-step 7f), otherwise be back to sub-step 7b).

7f) the first row data of information matrix V are carried out to calculus of differences, obtain difference information vector y,

y＝[y ₁,…,y _s,…,y _K-1] ^T

y _s＝v _1s+1-v _1s

Wherein, s=1,2 ..., K-1, v _1srepresent the 1st row s column element of information matrix V, y _ss the component that represents difference information vector y, subscript T represents vectorial transposition.

Whether the absolute value that 7g) judges all elements of difference information vector y is all less than M, if so, obtains angle on target index value S, then carry out sub-step 7k), wherein v _1ifor information matrix V the 1st row i column element; If the absolute value of all elements of difference information vector y is not all less than M, carry out sub-step 7h).

7h) effect angular range [the θ in azimuth dimension by radar _d, θ _u] be uniformly-spaced divided into H angular interval, wherein H is greater than 1 positive integer, the coboundary β of h angular interval _h? _uwith lower boundary β _h? _dbe respectively:

${\mathrm{\&beta;}}_{\mathrm{h\; u}}={\mathrm{\&theta;}}_d+\frac{({\mathrm{\&theta;}}_u-{\mathrm{\&theta;}}_d)}{H}h$

${\mathrm{\&beta;}}_{\mathrm{h\; d}}={\mathrm{\&theta;}}_d+\frac{({\mathrm{\&theta;}}_u-{\mathrm{\&theta;}}_d)}{H}(h-1)$

Wherein, h=1,2 ..., H.

7i) in statistical information matrix V with h the element number μ that angular interval is corresponding _h, μ _hfor:

${\mathrm{\&mu;}}_h=\underset{p=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{K}{\mathrm{\&Sigma;}}}{\mathrm{\&delta;}}_{\mathrm{pi}}$

Wherein,

Wherein, v _p? _ifor the capable i column element of p of information matrix V, Δ θ is the angle searching precision of radar pitching dimension.

The subscript t of the angular interval that then, in information matrix V, element number is maximum:

$t=\underset{h}{\arg \max }{\mathrm{\&mu;}}_h.$

7j) the element distribution centroid M' of computing information matrix V,

$M^{\&prime;}=\underset{w=t-2}{\overset{t+2}{\mathrm{\&Sigma;}}}w{\mathrm{\&mu;}}_w/\underset{w=t-2}{\overset{t+2}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_w$

Wherein, μ _wfor in information matrix V with w the element number that angular interval is corresponding, obtain angle on target index value S, S=M'.

7k), by angle on target index value S, obtain target elevation measured value

Effect of the present invention can further illustrate by following emulation experiment:

1) simulated conditions:

In emulation experiment of the present invention, software emulation platform is MATLAB R2010a, sets the thunder method wavelength X=2m that transmits, element number of array N=8 in array antenna, array element distance d=λ/2, the high H=25m of array antenna frame in experiment; Radar pitching dimension spatial domain effect angular range is [0.5 °, 10 °], and angle searching precision is Δ θ=0.05 °; For the information frame number K=5 of information fusion, multipath reflection coefficient ρ=-0.9.

2) emulation content and result:

Emulation 1, angle measurement deviation when this experiment is measured for low elevation angle target luffing angle for analyzing the inventive method and traditional RML method, under these conditions, sets the signal to noise ratio snr=5dB of single array element reception data, the true elevation angle theta of target ₀=3 ° and remain unchanged (in reality target elevation change very slow), utilize the inventive method and traditional RML method to carry out 100 secondary tracking measurements to angle on target, obtains two kinds of methods to the measurement result of target elevation as shown in Figure 3.With reference to Fig. 3, for adopting the present invention and traditional RML method angle on target to be carried out to the measurement result schematic diagram of the target elevation that 100 secondary trackings measure in emulation 1.In Fig. 3, transverse axis represents data frame number, and the longitudinal axis represents the measurement result of target elevation, and unit is degree.

As shown in Figure 3, in the measurement of angle result of traditional RML method, under some Frame, target elevation angle measurement is 6 ° and 9 ° of left and right, this is with 3 °, the true elevation angle of the target very large measured deviation of existence, and in the inventive method, the elevation angle angle measurement under all Frames is all very little with the measured deviation between the true elevation angle of target.

Emulation 2, when this experiment is measured low elevation angle angle on target for simulation analysis the inventive method and traditional RML, angle measurement root-mean-square error is with the situation of change of snr of received signal, under these conditions, the true angle θ of target setting ₀=3 °, the signal to noise ratio snr that single array element receives data changes to 20dB by 0dB, under each signal to noise ratio (S/N ratio), adopt the inventive method and traditional RML method to carry out independent measurement 1000 times to target azimuth angle, the angle measurement root-mean-square error that obtains two kinds of methods with the situation of change of signal to noise ratio (S/N ratio) as shown in Figure 4.With reference to Fig. 4, for adopting the present invention and traditional RML method to carry out angle measurement root-mean-square error that 1000 independent measurements the draw variation relation schematic diagram with signal to noise ratio (S/N ratio) to target azimuth angle in emulation 2.In Fig. 4, transverse axis represents signal to noise ratio (S/N ratio), and unit is dB, and the longitudinal axis represents angle measurement root-mean-square error, and unit is degree.

As shown in Figure 4, the inventive method has the better angle measurement accuracy of more traditional RML method, and particularly, in the situation that snr of received signal is lower, the advantage of the inventive method is more obvious.

In summary, the inventive method can be effective to the measurement of metric wave array radar to object height (elevation angle).

Obviously, those skilled in the art can carry out various changes and modification and not depart from the spirit and scope of the present invention the present invention.Like this, if these amendments of the present invention and within modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.

Claims (7)

1. the metric wave array radar target elevation measuring method based on multiframe information fusion, it is characterized in that, the array antenna of described radar is the even linear array being made up of N array element, and the described metric wave array radar target elevation measuring method based on multiframe information fusion comprises the following steps:

Step 1, draws the peak point number m in beam angle B and the radar beam scintigram of radar beam scintigram;

Step 2, determines the scouting interval comprising in the beam angle B of angle searching space-number L that radar comprises in the spatial domain of pitching dimension effect angular range and radar beam scintigram and counts M;

Step 3, utilizes radar receiving target echo data, obtains the echo data of K frame target place range unit, and wherein, the echo data of k frame target place range unit is expressed as x _k, k=1,2 ..., K;

Step 4, according to the array number N of radar emission signal wavelength lambda, radar array antenna, known multipath reflection coefficient ρ, radar acts on angular range lower bound θ in the spatial domain of pitching dimension _d, radar is at angle searching precision Δ θ and the high H of radar antenna frame of pitching dimension, draws beam scanning matrix W;

Step 5, utilizes beam scanning matrix W respectively the echo data of K frame target place range unit to be carried out to wave beam formation, and the wave beam that obtains each frame data forms vector, and wherein, it is z that the wave beam of the echo data of k frame target place range unit forms vector _k, by z ₁to z _kform beam forming matrix Z;

Step 6, extracts each wave beam and forms the sequence number of m data of amplitude maximum in vector, utilizes the sequence number extracting to form corresponding frame to receive the maximum value vector of data, and the maximum value vector of k frame reception data is v _k; By v ₁to v _kconfiguration information matrix V;

Step 7, carries out information fusion to information matrix V, obtains target elevation measured value θ.

2. the metric wave array radar target elevation measuring method based on multiframe information fusion as claimed in claim 1, is characterized in that, the concrete sub-step of described step 1 is:

1a) radar emission signal wavelength is λ, and radar antenna frame height is H, draws the beam angle B of radar beam scintigram, B=50/ (H/ λ+N);

1b) radar is [θ at the spatial domain of pitching dimension effect angular range _d, θ _u], at the spatial domain of pitching dimension effect angular range, draw the peak point number m obtaining in radar beam scintigram by radar, wherein expression rounds up.

3. the metric wave array radar target elevation measuring method based on multiframe information fusion as claimed in claim 1, is characterized in that, the concrete sub-step of described step 2 is:

2a) radar is [θ at the spatial domain of pitching dimension effect angular range _d, θ _u], the angle searching precision Δ θ being tieed up by radar pitching, obtains radar at the spatial domain of pitching dimension effect angular range [θ _d, θ _u] in the angle searching space-number L that comprises,

2b) the angle searching precision Δ θ in pitching dimension by radar, obtains the angle searching space-number M that beam angle B comprises,

4. the metric wave array radar target elevation measuring method based on multiframe information fusion as claimed in claim 1, is characterized in that, the concrete sub-step of described step 4 is:

4a) represent scouting interval count value with γ, γ=1,2 ..., in the time of γ=1, carry out sub-step 4b);

4b) in the bay of radar array antenna, an optional array element is as with reference to array element; By radar emission signal wavelength lambda and bay number N, structure weight vector w _{γ 1}with weight vector w _{γ 2}, wherein,

$w_{\mathrm{\&gamma;}1}=[e^{j2\mathrm{\&pi;}\frac{d_1}{\mathrm{\&lambda;}}\left[\sin ((\mathrm{\&gamma;}-1)\mathrm{\&Delta;\&theta;}-{\mathrm{\&theta;}}_d)\right]},...,e^{j2\mathrm{\&pi;}\frac{d_n}{\mathrm{\&lambda;}}\left[\sin ((\mathrm{\&gamma;}-1)\mathrm{\&Delta;\&theta;}-{\mathrm{\&theta;}}_d)\right]},...,e^{j2\mathrm{\&pi;}\frac{d_N}{\mathrm{\&lambda;}}\left[\sin ((\mathrm{\&gamma;}-1)\mathrm{\&Delta;\&theta;}-{\mathrm{\&theta;}}_d)\right]}{]}^T$

$w_{\mathrm{\&gamma;}2}=[e^{j2\mathrm{\&pi;}\frac{d_1}{\mathrm{\&lambda;}}\left[\sin ((\mathrm{\&gamma;}-1)\mathrm{\&Delta;\&theta;}-{\mathrm{\&theta;}}_d)\right]},...,e^{j2\mathrm{\&pi;}\frac{d_n}{\mathrm{\&lambda;}}\left[\sin ((\mathrm{\&gamma;}-1)\mathrm{\&Delta;\&theta;}-{\mathrm{\&theta;}}_d)\right]},...,e^{j2\mathrm{\&pi;}\frac{d_N}{\mathrm{\&lambda;}}\left[\sin ((\mathrm{\&gamma;}-1)\mathrm{\&Delta;\&theta;}-{\mathrm{\&theta;}}_d)\right]}{]}^{T,}$

Wherein, j is imaginary unit, d _nfor the distance between n array element and the reference array element of radar array antenna, n=1,2 ..., N; The transposition of subscript T representing matrix or vector; Δ θ is the angle searching precision of radar in pitching dimension, θ _dfor radar is at the lower bound of the spatial domain of pitching dimension effect angular range;

4c), by the high H of radar array antenna holder, obtain the phase differential between direct-path signal and multipath signal for:

40) by known multipath reflection coefficient ρ, weight vector w _{γ 1}with weight vector w _{γ 2}, obtain broadcasting bundle scanning weight vector w _γ, w _γfor:

4e) make the value of γ from increasing 1, and judge that whether scouting interval counter γ is greater than L, if so, obtains beam scanning matrix W, W=[w ₁, w ₂..., w _l], then perform step 5; If γ is not more than L, be back to sub-step 4b).

5. the metric wave array radar target elevation measuring method based on multiframe information fusion as claimed in claim 1, is characterized in that, the concrete sub-step of described step 5 is:

5a) represent frame count value with k, k=1,2 ..., in the time of k=1, carry out sub-step 5b);

5b) utilize the echo data x of beam scanning matrix W to k frame target place range unit _kcarry out beam scanning, obtain wave beam and form vector z _k, z _k=W ^hx _k, wherein, the conjugate transpose of subscript H representing matrix;

5c) make the value of k from increasing 1, and judge whether k is greater than K, if so, obtain beam forming matrix Z, Z=[z ₁, z ₂..., z _k], then perform step 6; If k is not more than K, be back to sub-step 5b).

6. the metric wave array radar target elevation measuring method based on multiframe information fusion as claimed in claim 1, is characterized in that, the concrete sub-step of described step 6 is:

6a) represent frame count value with k, k=1,2 ..., in the time of k=1, carry out sub-step 6b);

6b) represent peak point number count value with η, η=1,2 ..., in the time of η=1, carry out sub-step 6c);

6c) find out wave beam and form vector z _kthe sequence number I of the data of middle amplitude maximum _η, by I _ηreceive data maximum value vector v as k frame _kη component v _{η k}, wave beam is formed to vector z _kin with I _η2M+1 centered by individual data data value zero setting, wherein, M is the angle searching space-number that beam angle B comprises;

6d) make the value of η from increasing 1, judge whether η is greater than m, if so, obtain k frame and receive data maximum value vector v _k, v _k=[v _1k, v _2k..., v _mk] ^t, then carry out sub-step 6e); If η is not more than m, be back to sub-step 6c);

The value that 6e) makes k is from increasing 1, and whether k be greater than K, if so, obtains information matrix V, V=[v ₁, v ₂..., v _k], then perform step 7; If k is not more than K, be back to sub-step 6b).

7. the metric wave array radar target elevation measuring method based on multiframe information fusion as claimed in claim 1, is characterized in that,

7a) represent loop count with l, l=1,2 ..., in the time of l=1, carry out sub-step 7b);

7b) the degrees of fusion matrix A of computing information matrix V, degrees of fusion matrix A is the matrix of K × K dimension, when natural number p gets the arbitrary numerical value in 1 to m, all has | v _pi-v _pk| when≤N, the capable k column element of the i a of degrees of fusion matrix A _ikbe 1; Otherwise the capable k column element of the i a of degrees of fusion matrix A _ikbe 0; Wherein, i=1,2 ..., K, k=1,2 ..., K, v _pifor the capable i column element of p of information matrix V, || represent to take absolute value;

Whether all elements that 7c) judges degrees of fusion matrix A all equals 1, if so, obtains the angle index value S of target, wherein v _1ifor information matrix V the 1st row i column element, then carry out sub-step 7k); If all elements of degrees of fusion matrix A is not 1 entirely, carry out sub-step 7d);

7d) find out in degrees of fusion matrix A, the rower r of the row that nonzero element number is maximum,

$r=\underset{i}{\arg \max }\mathrm{sum}\left(a_i\right)$

Wherein, a _ifor the vector of the capable composition of i of degrees of fusion matrix A, sum () represent to ask each element in vector and;

Then, upgrade degrees of fusion matrix A, the process of upgrading degrees of fusion matrix A is: the vectorial a that judges the capable composition of r of degrees of fusion matrix A _rq component whether be 0, q=1,2 ..., K; If the vectorial a of the capable composition of the r of degrees of fusion matrix A _rq component be not 0, keep the q column data of information matrix V constant; If the vectorial a of the capable composition of the r of degrees of fusion matrix A _rq component be 0, the q of lastest imformation matrix V row, the process of the q row of lastest imformation matrix V is: the q column data of information matrix V is carried out to ring shift, that is: works as j=1,2 ..., when m-1, make v _j? _q=v _j+1q; If j=m, makes v _j? _q=v _1q; Wherein, v _j? _qrepresent the capable q column element of j of information matrix V;

7e) make the value of l from increasing 1, judge whether l is greater than m, if so, carry out sub-step 7f), otherwise be back to sub-step 7b);

7f) the first row data of information matrix V are carried out to calculus of differences, obtain difference information vector y,

y＝[y ₁,…,y _s,…,y _K-1] ^T

y _s＝v _1s+1-v _1s

Wherein, s=1,2 ..., K-1, v _1srepresent the 1st row s column element of information matrix V, y _ss the component that represents difference information vector y, subscript T represents vectorial transposition;

Whether the absolute value that 7g) judges all elements of difference information vector y is all less than M, if so, obtains angle on target index value S, then carry out sub-step 7k), wherein v _1ifor information matrix V the 1st row i column element; If the absolute value of all elements of difference information vector y is not all less than M, carry out sub-step 7h);

7h) effect angular range [the θ in azimuth dimension by radar _d, θ _u] be uniformly-spaced divided into H angular interval, wherein H is greater than 1 positive integer, the coboundary β of h angular interval _huwith lower boundary β _hdbe respectively:

${\mathrm{\&beta;}}_{\mathrm{h\; u}}={\mathrm{\&theta;}}_d+\frac{({\mathrm{\&theta;}}_u-{\mathrm{\&theta;}}_d)}{H}h$

${\mathrm{\&beta;}}_{\mathrm{h\; d}}={\mathrm{\&theta;}}_d+\frac{({\mathrm{\&theta;}}_u-{\mathrm{\&theta;}}_d)}{H}(h-1)$

Wherein, h=1,2 ..., H;

7i) in statistical information matrix V with h the element number μ that angular interval is corresponding _h, μ _hfor:

${\mathrm{\&mu;}}_h=\underset{p=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{K}{\mathrm{\&Sigma;}}}{\mathrm{\&delta;}}_{\mathrm{pi}}$

Wherein,

Wherein, v _p? _ifor the capable i column element of p of information matrix V, Δ θ is the angle searching precision of radar pitching dimension;

The subscript t of the angular interval that then, in information matrix V, element number is maximum:

$t=\underset{h}{\arg \max }{\mathrm{\&mu;}}_h;$

7j) the element distribution centroid M' of computing information matrix V,

$M^{\&prime;}=\underset{w=t-2}{\overset{t+2}{\mathrm{\&Sigma;}}}w{\mathrm{\&mu;}}_w/\underset{w=t-2}{\overset{t+2}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_w$

Wherein, μ _wfor in information matrix V with w the element number that angular interval is corresponding, obtain angle on target index value S, S=M';

7k), by angle on target index value S, obtain target elevation measured value