CN103293517B - Diagonal-loading robust adaptive radar beam forming method based on ridge parameter estimation - Google Patents

Abstract

The invention discloses a diagonal-loading robust adaptive radar beam forming method based on ridge parameter estimation. According to an implementation scheme, the method includes: receiving signals in real time by radar antennas; constructing adaptive weight vectors of a radar; defining output power of a radar antennal array; adopting a new estimation criterion for estimating unknown vectors; utilizing the estimated values of the unknown vectors for acquiring loading amount; and acquiring adaptive weight vectors for forming beams. The loading mount is determined in an adaptive diagonal loading method in the prior art by means of applying the new estimation criterion and a convex optimization tool. The method has the advantages that accuracy of weighted vectors of the radar antennas is improved, beams narrow in major lobe, low in minor lobe and capable of pointing to a target direction are formed, distortion of radar beams is avoided, desired signal receiving of the radar is enhanced, receiving of interference and noise is inhibited, errors in radar applications are resisted, radar targets can detected accurately, and false report or failure in report of targets by the radar is avoided.

Description

Diagonal angle based on ridge parameter estimation loads sane adaption radar Beamforming Method

Technical field

The invention belongs to array signal process technique field, relate to the beam-forming technology of radar signal, specifically a kind of diagonal angle based on ridge parameter estimation loads sane adaption radar Beamforming Method, is mainly used in the detection of radar target.

Background technology

Array Signal Processing is an important branch of modern signal processing, and its application relates to multiple technical fields such as radar, communication, biomedical engineering and sonar.Wave beam forming is also called airspace filter, for detecting target in field of radar, being a main aspect of ARRAY PROCESSING, its essence is by being weighted airspace filter to each array element of radar antenna, reach the object strengthening radar target signal, suppress Radar jam signal.Adaptive beamformer can according to the weighting factor of each array element of change adaptively modifying radar of signal environment.Over nearly 30 years, a large amount of achievements in research has been emerged: sample matrix inversion (SMI) method for adaption radar Wave beam forming aspect, linear constraint minimal variance (LCMV) method, diagonal angle Loading Method, based on the adaption radar beam-forming schemes (HKB) of ridge regression, based on the covariance matrix revised law (GLC) etc. of generalized linear combination.

SMI method is the method for the conventional radar self-adaption ground detection of a target, and Project Realization is comparatively simple and easy.LCMV method is also the method for a kind of conventional radar self-adaption ground detection of a target, and it belongs to the category of SMI.In the practice of SMI method, various error can cause that the main lobe of radar beam offsets, minor level raises, radar beam will be caused time serious to distort, and the result caused is that radar can not the position of the accurately detection of a target, sometimes even detects fall short or false target.So the radar beam formation method being used for suppressing the diagonal angle of radar directional diagram distortion to load is arisen at the historic moment.Diagonal angle loading technique is: before inverting to radar reception data covariance matrix, to covariance matrix correction, and the method realizing revising loads the value that radar receives on data covariance matrix diagonal line.Diagonal angle loading technique can suppress the disturbance of little eigenwert and characteristic of correspondence vector, weakens the impact of radar noise, improves the distortion of radar directional diagram, and the robustness of anti-steering vector error is comparatively strong, meanwhile, can compression radar undesired signal, improve speed of convergence.

Traditional diagonal angle loads the heap(ed) capacity of adaption radar Beamforming Method by the radar noise power decision estimated, the constraint that the selection of heap(ed) capacity is unfixing, and normally radar user rule of thumb comes to select.Although this diagonal angle loading technique improves the distortion of radar directional diagram to a certain extent, still there is following problem: when one, heap(ed) capacity is too small, the improvement effect that radar directional diagram distorts is also not obvious; Two, when heap(ed) capacity is excessive, the inhibition strength of radar to interference can be reduced; Three, not only there is the impact of human factor in the heap(ed) capacity of this selection of the experience according to radar user; And can not change completely adaptively along with the change of the actual Received signal strength of radar, in the practical application of radar, there is significant limitation, in order to improve radar to the real-time search speed of target and accuracy, need a kind of Beamforming Method that can adjust radar beam according to the change of actual signal completely adaptively.

So someone proposes the Beamforming Method that can change completely adaptively according to the change of actual conditions: HKB method and GLC method.Wherein, determine that radar receives the diagonal loading amount of data the data adaptive that HKB method can receive according to radar, without the need to artificially selecting diagonal loading amount, and can good detection radar target when radar sampling data are few.But HKB method also has its weak point: the heap(ed) capacity in HKB method can increase along with the increase of fast umber of beats, and when fast umber of beats is larger, the excessive detection accuracy of HKB method that causes of heap(ed) capacity declines, radar detection fall short or false target time serious, will be caused.GLC method can receive data according to radar and directly find optimum steering vector or optimum radar data covariance matrix, thus improves the robustness of radar beam shaper.But, still there is many shortcomings in it: the model that the revaluation radar that, it adopts receives data covariance matrix is: the linear combination of radar data covariance matrix and unit matrix, namely suppose that the noise that radar receives is white Gaussian noise, if the noise that actual radar receives is coloured noise, this method just accurately can not estimate radar target; Two, in the process solved covariance matrix least mean-square error, due to the shortage of useful signal prior distribution information knowledge, estimate can there is comparatively big error in computing, this can cause the radar target error of detection larger; Three, can generate very little diagonal loading amount when fast umber of beats is larger, GLC method will reduce the robustness of anti-steering vector error, and namely when skew occurs antenna, GLC method accurate detection target even can not detect fall short.

In sum, carrying out in the scheme of radar target acquisition at existing Wave beam forming, there is radar beam distortion and change environmentally can not carry out the problem of the detection of radar target completely adaptively in SMI method and LCMV method; HKB method exist fast umber of beats slightly large time heap(ed) capacity excessive, radar data reduction decline or lost efficacy problem; The existence of GLC method loses the problems such as the robustness of opposing radar vectoring vector error to prior imformation error sensitive, snap a few hours.These problems make radar can not the detection of a target quickly and accurately, even occur the situation of detection fall short or false target.

Summary of the invention

The object of the invention is to for the deficiency in above-mentioned existing method, on the basis of existing adaption radar Beamforming Method, propose a kind of adaptive, heap(ed) capacity rational, the diagonal angle based on ridge parameter estimation that is insensitive to prior imformation error, that can resist steering vector error loads sane adaption radar Beamforming Method.

For achieving the above object, technical thought of the present invention is: the method directly extracts the information relevant to noise and error adaptively from the reception data of radar, use the information and new estimation criterion extracted, the radar more tallied with the actual situation fast receives the heap(ed) capacity of data covariance matrix, the weight vectors utilizing this heap(ed) capacity to obtain radar array carries out Adaptive beamformer, thus realize accurately detecting rapidly radar target, specifically comprise the steps:

Step 1) radar real-time reception signal, if radar return signal is x (t), carries out N real-time sampling to x (t), obtains the data x=[x (t of radar return signal ₁) ..., x (t _n)], radar receives the covariance matrix of data x to utilize maximum likelihood estimate to estimate wherein N is fast umber of beats.

Step 2) set the direction vector of radar target signal as a _s, according to a _sobtain the signal space B orthogonal with radar target signal space; Utilize B, constraint condition w ^ha _s=1 and unknown vector η construct radar weighing vector wherein, () ^hrepresent conjugate transpose, M is the array number of aerial array, and B is that M × (M-1) ties up complex matrix, and w is that complex vector is tieed up in M × 1, and η is that complex vector is tieed up in (M-1) × 1.

Step 3) radar beam formed model be:

$\underset{w}{\min }{w}^H\hat{R}w$ s.t.w ^Ha _s＝1

Use radar weighing vector obtain the objective function that radar beam is formed:

$\underset{w}{\min }{w}^H\hat{R}w=\underset{\mathrm{\η}}{\min }{(\mathrm{B\η}-\frac{a_s}{M})}^H\hat{R}(\mathrm{B\η}-\frac{a_s}{M})=\underset{\mathrm{\η}}{\min }{\left|\right|{\hat{R}}^{1/2}\mathrm{B\η}-{\hat{R}}^{1/2}\frac{a_s}{M}\left|\right|}_2^2$

Order e=b-X η is remainder vector, then the objective function that radar beam is formed is reduced to:

$\underset{\mathrm{\η}}{\min }{\left|\right|\mathrm{X\η}-b\left|\right|}_2^2$

Wherein, the value of w when expression gets minimum value, the value of η when expression gets minimum value, representing matrix on Square-Rooting Matrices, ‖ ‖ ₂represent 2 norms.

Step 4) with new estimation criterion estimating unknown vector η, then using standard interior point method to meeting estimation criterion unknown vector η search for, obtain estimator wherein, || || ₁represent 1 norm.In existing Beamforming Method, the estimation criterion used is all and the estimation criterion of the present invention to make unknown vector η according to actual conditions and more reasonably estimating, and then obtain more reasonably heap(ed) capacity, make the present invention possess stronger adaptivity.

Step 5) use step 4) in estimator data covariance matrix is received to radar loading coefficient ρ solve, wherein, ρ is a scalar.

Step 6) use radar to receive the covariance matrix of data diagonal angle loading coefficient ρ and the direction vector a of radar target signal _s, obtain the weighing vector w of radar antenna, the weighing vector w obtained sent to radar, upgrade the weighting coefficient of radar antenna, the signal of radar real-time reception forms after these coefficient weightings that main lobe is narrow, secondary lobe is low, can the wave beam in accurate sensing direction.Invention enhances the reception of radar to observed ray signal, inhibit radar to the reception of other direction incoming wave signals, reach the object of accurately detection radar target.

The present invention compared with prior art has the following advantages:

(1) the present invention has used new estimation criterion solve unknown vector η with convex optimization tool, obtain more realistic estimator then obtain the weight vectors of more accurate radar antenna array, the beam main lobe formed with this weight vectors be narrower, secondary lobe is lower, can sensing direction more accurately.In actual applications, fortune in this way, radar antenna array can be made to strengthen the reception of radar wanted signal, suppress the error that exists in the reception to radar jamming and noise, opposing radar system and prior imformation, reach the object of more effectively accurately detection radar target;

(2) the present invention carries out diagonal angle loading by covariance matrix radar being received to data, decrease the degree of scatter that radar receives data covariance matrix eigenwert, reduce main lobe width and the secondary lobe height of radar beam, avoid the distortion of radar beam, radar thus can be avoided in actual applications to make a false report or fail to report target;

(3) the present invention directly receives data from radar and extracts the information with environmental correclation adaptively, change environmentally can change adaptively, reaches the object of real-time detection radar target, have stronger adaptive ability.

Accompanying drawing explanation

Fig. 1 is realization flow figure of the present invention;

Fig. 2 is under ideal conditions, and the radar antenna array of distinct methods exports the curve map of Signal to Interference plus Noise Ratio with the fast umber of beats change of array received signal;

Fig. 3 is under non-ideal condition, and the radar antenna array of distinct methods exports the curve map of Signal to Interference plus Noise Ratio with the fast umber of beats change of array received signal;

Fig. 4 is the curve map that the radar antenna array output Signal to Interference plus Noise Ratio of distinct methods changes with array input signal-to-noise ratio;

Fig. 5 is the curve map that the radar antenna array output Signal to Interference plus Noise Ratio of distinct methods changes with array error in pointing.

Embodiment

Embodiment 1:

The present invention is that a kind of diagonal angle based on ridge parameter estimation loads sane adaption radar Beamforming Method, and for detection radar target, see the beam forming process of Fig. 1, specific embodiment of the invention comprises the steps:

Step 1: radar real-time reception signal, if radar return signal is x (t), carry out to x (t) the data x that N real-time sampling obtains radar return signal, radar receives the covariance matrix of data x to utilize maximum likelihood estimate to estimate solution procedure comprises the steps:

(1a) according to the structural model of radar antenna Received signal strength, be that the radar antenna of M is expressed as follows at radar return signal x (t) of t by array number:

$x\left(t\right)=s\left(t\right)+i\left(t\right)+n\left(t\right)=[a_s{a}_{i_1}\·\·\·a_{i_p}]\left[\begin{array}{c}s\left(t\right)\\ i_1\left(t\right)\\ \·\·\·\\ i_p\left(t\right)\end{array}\right]+n\left(t\right)$

Wherein, N is fast umber of beats, and p is the number of the undesired signal that radar receives, for the radar antenna array stream shape matrix that M × (p+1) ties up, a _sfor the steering vector of radar target signal s (t), be respectively Radar jam signal i ₁(t), i ₂(t) ..., i _pthe steering vector of (t), the noise signal that n (t) receives for radar, a _s, , n (t) is that complex vector is tieed up in M × 1;

(1b) N sampling is carried out to radar return signal x (t), obtains the reception data of radar antenna:

x＝[x(t ₁)，...，x(t _N)]

(1c) following formula is utilized to calculate the covariance matrix of radar antenna reception data x for:

$\hat{R}=\frac{1}{N}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}x\left(t_n\right)x^H\left(t_n\right)$

Wherein, () ^hrepresent conjugate transpose.

By step 1, obtain the real-time information of radar signal, from the real-time information obtained, next extract the information needed in the invention process process.

Step 2: the expression formula of the weighing vector w of structure radar antenna array:

(1a) set the direction vector of radar target signal as a _s, according to radar target sense vector a _s, solve and meet constraint condition: B ^ha _s=0, B ^hthe orthogonal complement space B of the radar target signal space of B=I; Wherein, () ^hrepresent conjugate transpose, B is that M × (M-1) ties up complex matrix;

(1b) according to the condition that the adaptive weighted vectorial w of orthogonal complement space B and radar antenna array should meet: w ^ha _s=1, the linear combination obtaining the weighing vector of radar antenna array is:

$w=\frac{a_s}{M}-\mathrm{B\η}$

Wherein, η ties up complex vector to be asked for (M-1) × 1, and M is the array number of aerial array.

Step 3: the model that radar beam is formed is:

$\underset{w}{\min }{w}^H\hat{R}w$ s.t.w ^Ha _s＝1

Use radar weighing vector obtain the objective function that radar beam is formed:

$\underset{w}{\min }{w}^H\hat{R}w=\underset{\mathrm{\η}}{\min }{(\mathrm{B\η}-\frac{a_s}{M})}^H\hat{R}(\mathrm{B\η}-\frac{a_s}{M})=\underset{\mathrm{\η}}{\min }{\left|\right|{\hat{R}}^{1/2}\mathrm{B\η}-{\hat{R}}^{1/2}\frac{a_s}{M}\left|\right|}_2^2$

Order e=b-X η is remainder vector, then the objective function that radar beam is formed is reduced to:

$\underset{\mathrm{\η}}{\min }{\left|\right|\mathrm{X\η}-b\left|\right|}_2^2$

Wherein, the value of w when expression gets minimum value, the value of η when expression gets minimum value, s.t. represents that constraint condition is, representing matrix on Square-Rooting Matrices, || || ₂represent 2 norms.

Step 4: with new estimation criterion estimating unknown vector η, then using standard interior point method to meeting estimation criterion unknown vector η search for, concrete method for solving is:

Use the standard interior point method software CVX in ripe convex optimization field, to meeting objective function the unknown vector η of radar search for, obtain the estimator of η wherein, || || ₁represent 1 norm.

Step 5: use the estimator in step 4 radar is received to the covariance matrix of data loading coefficient ρ solve, concrete method for solving is:

By the estimator in step 4 be updated to the covariance matrix that radar receives data the expression formula of diagonal angle loading coefficient in, obtain diagonal angle loading coefficient ρ, wherein || || ₂represent 2 norms.

Step 6: use radar to receive the covariance matrix of data diagonal angle loading coefficient ρ and the direction vector a of radar target signal _s, obtain the weighing vector w of radar antenna, concrete method for solving is:

(6a) radar is used to receive the covariance matrix of data diagonal angle loading coefficient ρ and the direction vector a of radar target signal _s, obtain the weighing vector of radar antenna array

$w=\frac{{(\hat{R}+\mathrm{\ρI})}^{-1}{a}_s}{{a_s}^H(\hat{R}+\mathrm{\ρI})a_s}$

Wherein, () ^hrepresent conjugate transpose, () ^-1representing matrix is inverted, and I is that M ties up unit matrix.

(6b) the weighing vector w obtained in step (6a) is sent to radar, upgrade the weighting coefficient of radar antenna, the signal of radar real-time reception forms after these coefficient weightings that main lobe is narrow, secondary lobe is low, can the wave beam in accurate sensing direction, enhance the reception of radar to observed ray signal, inhibit radar to the reception of other direction incoming wave signals, reach the object of accurately detection radar target.

Embodiment 2:

Diagonal angle based on ridge parameter estimation loads sane adaption radar Beamforming Method with embodiment 1, and effect of the present invention further illustrates as follows in conjunction with emulation experiment:

Emulation 1: export Signal to Interference plus Noise Ratio and fast umber of beats Relationship Comparison:

Simulated conditions: the even linear array of radar antenna array model to be spacing be half-wavelength, array number is 10, and the number of radar target signal is 1, and the number of undesired signal is 2, and signal to noise ratio snr is 0dB, and dry making an uproar compares INR ₁=INR ₂=10dB, the arrival bearing of radar target signal is θ _s=20 °, the arrival bearing of undesired signal 1 is θ _i1=-30 °, the spatial frequency of undesired signal 2 direction vector is wherein, γ=0.9, Monte Carlo number of times is 200 times.

Emulation content:

Under ideal conditions, namely when there is no radar vectoring vector error, export Signal to Interference plus Noise Ratio by existing conventional beamformer method, HKB beam-forming schemes, generalized linear combination GLC robust ada-ptive beamformer method and method of the present invention to radar antenna array to emulate with the fast umber of beats change of radar antenna array Received signal strength, simulation result as shown in Figure 2.As can be seen from Figure 2, the comparatively existing Beamforming Method of the present invention has had and has significantly improved on detection performance, and this is the precision owing to invention increases radar antenna weight vectors.In actual applications, can strengthen radar antenna array to the reception of radar wanted signal, suppress to radar jamming and noise reception, reach the object of more effectively accurately detection radar target.

Embodiment 3:

Diagonal angle based on ridge parameter estimation loads sane adaption radar Beamforming Method with embodiment 1, and effect of the present invention further illustrates as follows in conjunction with emulation experiment:

Emulation 2: export Signal to Interference plus Noise Ratio and fast umber of beats Relationship Comparison:

Simulated conditions: radar vectoring vector error other simulated conditions are identical with emulation 1.

Emulation content: under non-ideal condition, namely when there is steering vector error in radar, export Signal to Interference plus Noise Ratio by existing conventional beamformer method, HKB beam-forming schemes, generalized linear combination GLC robust ada-ptive beamformer method and method of the present invention to radar antenna array to emulate with the change of radar antenna array input signal-to-noise ratio, simulation result as shown in Figure 3.As can be seen from Figure 3, under non-ideal condition, other algorithm performances decline obviously, and the present invention still can have good performance, illustrates that the present invention has the robustness of stronger opposing radar vectoring vector error.

Embodiment 4:

Diagonal angle based on ridge parameter estimation loads sane adaption radar Beamforming Method with embodiment 1, and effect of the present invention further illustrates as follows in conjunction with emulation experiment:

Emulation 3: export Signal to Interference plus Noise Ratio and input signal-to-noise ratio Relationship Comparison:

Simulated conditions: suppose that fast umber of beats N=20, SNR change, other simulated conditions are identical with emulation 1.

Emulation content: export Signal to Interference plus Noise Ratio by existing conventional beamformer method, HKB beam-forming schemes, generalized linear combination GLC robust ada-ptive beamformer method and method of the present invention to radar antenna array and emulate with the change of radar antenna array input signal-to-noise ratio, simulation result as shown in Figure 4.As can be seen from Figure 4, the present invention carries out the loading of self-adaptation diagonal angle by covariance matrix radar being received to data, avoids the distortion of radar beam, significantly improves than additive method performance.

Embodiment 5:

Diagonal angle based on ridge parameter estimation loads sane adaption radar Beamforming Method with embodiment 1, and effect of the present invention further illustrates as follows in conjunction with emulation experiment:

Emulation 4: export Signal to Interference plus Noise Ratio and array error in pointing Relationship Comparison:

Simulated conditions: suppose fast umber of beats N=20, radar array error in pointing is θ _error=-2 ° ~ 2 °, other simulated conditions are identical with emulation 1.

Emulation content: export Signal to Interference plus Noise Ratio by existing conventional beamformer method, HKB beam-forming schemes, generalized linear combination GLC robust ada-ptive beamformer method and the inventive method to radar antenna array and emulate with the change of radar antenna array error in pointing, simulation result as shown in Figure 5.As can be seen from Figure 5, the present invention has the ability of stronger opposing prior imformation error than additive method; Meanwhile, when fast umber of beats is only 20, the present invention still has good performance, illustrates that the present invention can change in real time in change environmentally, reaches the object of real-time detection radar target, have stronger adaptive ability.

In sum, the diagonal angle based on ridge parameter estimation of the present invention loads sane adaption radar Beamforming Method, and implementation comprises: radar antenna real-time reception signal; Structure radar self-adaption weight vector; Definition radar antenna array output power; New cost function is adopted to estimate unknown vector; Unknown vector estimated value is utilized to obtain heap(ed) capacity; Be obtained from adaptation weight vector, carry out Wave beam forming.The present invention has used new estimation criterion and convex optimization tool to solve the problem identificatioin of heap(ed) capacity in self-adaptation diagonal angle loading method in prior art.Invention increases the precision of radar antenna weight vectors, define that main lobe is narrow, secondary lobe is low, can point to the wave beam of target direction adaptively, avoid the distortion of radar beam, enhance radar to the reception of wanted signal, inhibit the reception to interference and noise, resisted the error existed in radar application, can detection radar target more accurately, avoid radar make a false report or fail to report target.

Claims (3)

1. the diagonal angle based on ridge parameter estimation loads a sane adaption radar Beamforming Method, for detection radar target, it is characterized in that: comprise the steps:

Step 1) set radar return signal as x (t), carry out to x (t) the data x that N real-time sampling obtains radar return signal, radar receives the covariance matrix of data x to utilize maximum likelihood estimate to estimate wherein N is fast umber of beats;

Step 2) set the direction vector of radar target signal as a _s, according to a _sobtain the space B orthogonal with radar target signal space; Utilize B and constraint condition w ^ha _s=1, by the linear combination expression formula of radar weighing vector w with unknown vector η represent, wherein, () ^hrepresent conjugate transpose, M is the array number of aerial array;

Step 3) the debatable model of radar beam shape is:

$\begin{array}{cc}\underset{w}{\min }{w}^H\hat{R}w& s.t.w^H{a}_s=1\end{array}$

Use radar weighing vector obtain the debatable objective function of radar beam shape:

$\underset{w}{\min }{w}^H\hat{R}w=\underset{\mathrm{\η}}{\min }{(\mathrm{B\η}-\frac{a_s}{M})}^H\hat{R}(\mathrm{B\η}-\frac{a_s}{M})=\underset{\mathrm{\η}}{\min }{\left|\right|{\hat{R}}^{1/2}\mathrm{B\η}-{\hat{R}}^{1/2}\frac{a_s}{M}\left|\right|}_2^2$

Order e=b-X η is remainder vector, then the debatable objective function of radar beam shape is reduced to:

$\underset{\mathrm{\η}}{\min }{\left|\right|\mathrm{X\η}-b\left|\right|}_2^2$

Wherein, the value of w when expression gets minimum value, the value of η when expression gets minimum value, s.t. represents that constraint condition is, representing matrix on Square-Rooting Matrices, || || ₂represent 2 norms;

Step 4) with new estimation criterion estimating unknown vector η, then using standard interior point method to meeting estimation criterion unknown vector η search for, obtain estimator wherein, || || ₁represent 1 norm;

Step 5) use estimator obtain the covariance matrix that radar receives data diagonal angle loading coefficient ρ, wherein ρ is a scalar; By estimator apply to radar and receive data covariance matrix the expression formula of diagonal angle loading coefficient in, obtain diagonal angle loading coefficient ρ, wherein, M is the array number of aerial array, || || ₂represent 2 norms;

Step 6) use radar to receive the covariance matrix of data diagonal angle loading coefficient ρ and the direction vector a of radar target signal _s, obtain the weighing vector w of radar antenna, the weighing vector w obtained is sent to radar, upgrade the weighting coefficient of radar antenna, the signal of radar real-time reception forms the wave beam in sensing direction after these coefficient weightings.

2. a kind of diagonal angle based on ridge parameter estimation according to claim 1 loads sane adaption radar Beamforming Method, it is characterized in that: wherein step 1) in: the covariance matrix of radar real-time reception data x and x solution procedure comprise the steps:

(1a) according to the structural model of radar antenna Received signal strength, be that the radar antenna of M is expressed as follows at Received signal strength x (t) of t by array number:

$x\left(t\right)=s\left(t\right)+i\left(t\right)+n\left(t\right)=\left[\begin{array}{cccc}{a}_s& a_{i_1}& ...& a_{i_p}\end{array}\right]\left[\begin{array}{c}s\left(t\right)\\ i_1\left(t\right)\\ ...\\ i_p\left(t\right)\end{array}\right]+n\left(t\right)$

Wherein, p is the number of the undesired signal that radar receives, $A=\left[\begin{array}{cccc}{a}_s& a_{i_1}& ...& a_{i_p}\end{array}\right]$ For the radar antenna array flow pattern matrix that M × (p+1) ties up, a _sfor the steering vector of radar target signal s (t), be respectively Radar jam signal i ₁(t), i ₂(t) ..., i _pthe steering vector of (t), the noise signal that n (t) receives for radar, n (t) is that complex vector is tieed up in M × 1;

(1b) N sampling is carried out to radar return signal x (t), obtains the reception data of radar antenna:

x＝[x(t ₁)，...，x(t _N)]

(1c) maximum likelihood estimate is utilized to calculate the covariance matrix of radar antenna reception data x:

$\hat{R}=\frac{1}{N}\underset{n=1}{\overset{N}{\mathrm{\Σ}}}x\left(t_n\right)x^H\left(t_n\right)$

Wherein, N represents fast umber of beats, () ^hrepresent conjugate transpose.

3. the diagonal angle based on ridge parameter estimation according to claim 2 loads sane adaption radar Beamforming Method, it is characterized in that: wherein step 6) in: use radar to receive the covariance matrix of data diagonal angle loading coefficient ρ and the direction vector a of radar target signal _s, obtain the weighing vector w of radar antenna, concrete method for solving is:

Radar is used to receive the covariance matrix of data diagonal angle loading coefficient ρ and the direction vector a of radar target signal _s, obtain the weighing vector of radar antenna array w is applied in radar antenna and carry out Wave beam forming, realize the detection to radar target, wherein, () ^hrepresent conjugate transpose, () ^-1representing matrix is inverted, and I is that M ties up unit matrix.