Anti-jamming method based on data association of active / passive radar compound seeker

Abstract

The present invention discloses an anti-jamming method based on data association of a primary / passive radar combined seeker, which mainly solves the problems of large computational amount, high equipment cost and great influence by environmental factors in the prior art. The implementation scheme is as follows: the primary / passive radars respectively perform tracking filtering on the target to obtain state estimation values and error covariance matrices; the state estimation values and error covariance matrices of the active radar are converted into the polar coordinate system; track association is performed on the filtering data of the primary / passive radars in the polar coordinate system; the types of jamming are identified according to the track association results and decision conditions; when not being jammed, the data after primary / passive fusion is used for guidance, and when being jammed, the data of the passive radar is used for guidance. The present invention can effectively identify the types of jamming, reduce the probability of the seeker being deceived, improve the tracking accuracy of the target, and can be used for the tracking guidance of a primary / passive combined system seeker.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar, and further relates to an anti-jamming method, which can be used for a seeker to track a ship target. Background Art

[0002] Centroid passive jamming and outboard active jamming are the most common jamming measures for ships to counter anti-ship missiles. The mechanisms of the two jammings are relatively similar, both causing the seeker to track the energy centroid of the target and the jamming in the terminal guidance stage; the jamming echo of centroid passive jamming is mainly generated by the reflection of the radar transmission signal by chaff and other jamming objects, and the jamming echo of outboard active jamming is independently generated by the jammer. For resisting these jammings, there are still certain problems and deficiencies in the existing technologies, such as complex calculation, too high cost or being greatly affected by the environment, etc.

[0003] Liang Ziyao proposed a passive jamming discrimination method based on polarization characteristics in the paper "Research on Polarization Anti-Jamming Technology of Anti-Ship Radar Seeker" (China Academy of Launch Vehicle Technology, master's thesis, February 2022). The specific steps of this method are: the first step, the polarization mode of the radar transmission signal is periodically switched between "vertical" and "horizontal", and at the same time, the reflected echoes of horizontal and vertical polarizations are received; the second step, after the received echo signals are processed by the signal processing module through digital filtering, digital pulse compression, coherent accumulation, constant false alarm detection, target clustering, etc., the time, frequency, and polarization domain characteristic parameters of the target are extracted; the third step, calculate the horizontal and vertical polarization ratios. The polarization ratios in the horizontal and vertical directions of the real target are both large and the energies are close, while the difference in the horizontal and vertical polarization ratios of the jamming is large. Thus, the presence of passive jamming can be discriminated. Since this method has a complex working mode and needs to extract the time, frequency, and polarization domain characteristic parameters of the target in real time, the calculation amount of the signal processing module is increased, and the manufacturing difficulty and cost are increased.

[0004] Li Gang et al. proposed an anti-jamming tracking method for a radar / infrared composite seeker in the paper "Anti-Jamming Tracking Method for Radar / Infrared Composite Seeker" (Flight Dynamics, Vol. 34, No. 5, 2016). The specific steps of this method are: the first step, perform correlation detection on the observation information of the target obtained by the radar and infrared sensors. If the correlation is high, it is considered that the seeker is not effectively jammed. If the correlation is low, it is considered that the seeker is jammed; the second step, if the seeker is not jammed, use the distributed fusion algorithm to estimate the target state, otherwise detect the trace value of the filtering innovation variance of each sensor, and consider that the sensor corresponding to the larger trace value is jammed; the third step, the system outputs the target state estimate of the sensor corresponding to the smaller trace value and uses it for guidance. Since this method uses an infrared sensor to obtain the observation information of the target, it not only has a high cost, but also the detection range is seriously affected by weather and environment. At the same time, there are also requirements for the flight speed of the seeker, and the limitations are strong. Summary of the Invention

[0005] Aiming at the deficiencies of the above-mentioned existing technologies, the purpose of the present invention is to propose an anti-jamming method based on a combined active / passive radar seeker, so as to reduce the computational complexity of interference recognition, improve the environmental adaptability and reliability of the equipment, and reduce the use cost of the equipment.

[0006] The technical solution of the present invention is as follows: anti-jamming is achieved by combining the detection data of active radar and passive radar, that is, the detection data of active radar is converted to the rectangular coordinate system for filtering; the detection data of passive radar is directly filtered in the polar coordinate system; then the state estimate and error covariance of active radar are converted to the polar coordinate system for correlation processing; then, according to the characteristics of different interferences, decision conditions are added to identify the types of interferences; finally, the state of the target is estimated and the estimate is output. The implementation steps are as follows:

[0007] (1) Obtain the measurement data of the active / passive radar from the position information of the ship target, seeker and interference, filter it, and calculate the state estimate of the active radar in the rectangular coordinate system Error covariance matrix P k,a and the state estimate of the passive radar in the polar coordinate system Error covariance matrix P k,p ;

[0008] (2) According to the state estimate of the active radar in the rectangular coordinate system and the error covariance P k,a , calculate the state estimate of the active radar in the polar coordinate system and the error covariance

[0009] (3) According to the state estimate of the active radar in the polar coordinate system Error covariance matrix and the state estimate of the passive radar in the polar coordinate system Error covariance matrix P k,p , calculate the Mahalanobis distance d between the state estimates at each moment during the filtering process k,ap ;

[0010] (4) According to the Mahalanobis distance d k,ap calculate the track association statistic λ at each moment during the filtering process of the active / passive radar k,ap ;

[0011] (5) According to the track association statistic λ k,ap , use hypothesis testing to obtain the track association result of the active / passive radar;

[0012] (6) Identify the types of interferences according to the track association result;

[0013] (6a) Identify whether the seeker is interfered according to the association result between the active track and the passive track of the target at time k:

[0014] If the association is successful, it means the seeker is not interfered, and step (7) is executed;

[0015] Otherwise, it means the seeker is interfered, and step (6b) is executed;

[0016] (6b) Identify the type of interference on the seeker according to the association result between the active track of the target and the suspicious passive track at time k:

[0017] If the association is successful, it means the seeker is interfered by an external active interference, and step (8) is executed;

[0018] Otherwise, it means the seeker is interfered by a centroid passive interference, and step (8) is executed;

[0019] (7) According to the state estimator of the active radar error covariance matrix and the state estimator of the passive radar error covariance matrix P k,p , calculate the state estimator through track fusion and use it for the guidance of the seeker;

[0020] (8) Guide the seeker according to the state estimator of the passive radar

[0021] The present invention has the following advantages compared with the prior art:

[0022] First, since the present invention combines the detection data of the main / passive radars at the data processing layer to identify interference, it only depends on the time-domain information of the target, does not need to extract the target frequency-domain and polarization-domain information, and does not need to decompose the polarization characteristics through multi-pulse accumulation. The calculation is simple, overcoming the defects of complex decision conditions and large calculation amount in the prior art, and significantly improving the interference recognition efficiency of the present invention.

[0023] Second, since the present invention adopts the cooperation mode of the main / passive radars, the equipment cost is low, and it is less affected by weather and environment, overcoming the deficiencies of the prior art that the detection ability is seriously affected by weather and environment and the flight speed of the carrying platform is limited. The present invention can stably perform long-distance detection and can be applied to high-speed platforms, significantly improving the performance-price ratio and versatility of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is the implementation flowchart of the present invention;

[0025] Figure 2 It is the deviation curve graph between the filtering value and the true value of the azimuth angle of the main / passive radar target under the centroid passive interference;

[0026] Figure 3 It is the change curve graph of the track association statistic of the present invention under the centroid passive interference;

[0027] Figure 4 It is the deviation curve graph between the filtering value and the true value of the azimuth angle of the main / passive radar target under the outboard active interference;

[0028] Figure 5 It is the change curve graph of the track association statistic of the present invention under the outboard active interference. Specific implementation manner

[0029] The following further describes the embodiments and effects of the present invention in conjunction with the accompanying drawings.

[0030] Refer to Figure 1 , and the implementation steps of this example are as follows:

[0031] Step 1, obtain the measurement data of the main / passive radar from the position information of the ship target, seeker and interference.

[0032] In this step, corresponding measurement data is generated according to different interference types, and its implementation is as follows:

[0033] Suppose the position xyz of the seeker at time k in the three-dimensional plane k,m =(x k,m , y k,m , z k,m ), and the position xyz of the target or equivalent centroid k,t =(x k,t , y k,t , z k,t ), then its polar coordinate information relative to the seeker is:

[0034]

[0035] Among them, respectively represent the radial distance, azimuth angle, and pitch angle;

[0036] (1.1) Generate measurement data without interference:

[0037] Regard each radar and seeker as the same mass point. Before the interference is released, both the active radar and the passive radar will track the target;

[0038] Suppose the position xyz of the target k,s =(x k,s , y k,s , z k,s) According to the characteristics that an active radar can obtain the radial distance, azimuth angle, and elevation angle information of a target, while a passive radar can only obtain the azimuth angle and elevation angle information, substitute xyz k,t =xyz k,s into Equation <1> to calculate the measurement value Z of the active radar respectively k,a and the measurement value Z of the passive radar k,p :

[0039]

[0040] where are the measurement values of the active radar for the target's radial distance, azimuth angle, and elevation angle respectively, are the measurement values of the passive radar for the target's azimuth angle and elevation angle respectively, and W k,a is the observation noise of the active radar, and W k,p is the observation noise of the passive radar;

[0041] (1.2) Generate measurement data under interference:

[0042] 1.2.1) Generate measurement data under centroid passive interference:

[0043] Based on the fact that after the interference is released, the passive radar will continue to track the target, while the active radar will track the equivalent centroid of the target and the interference. Assume the position of the target is xyz k,s =(x k,s , y k,s , z k,s ), the position of the interference is xyz k,j1 =(x k,j1 , y k,j1 , z k,j1 ), and the position of the centroid is xyz k,c1 =(x k,c1 , y k,c1 , z k,c1 ). Before the seeker flies to the resolvable distance, calculate the position information of the centroid:

[0044]

[0045] where σ s , σ j are the equivalent radar cross-sections of the target and the interference respectively;

[0046] Let xyz k,t =xyz k,c1 , and combine Equation <1> and Equation <2> to calculate the measurement value Z of the active radar for the centroid under passive interference k,a1 :

[0047]

[0048] Among them, are the measured values of the equivalent centroid's radial distance, azimuth angle, and pitch angle by the active radar, respectively;

[0049] 1.2.2) Generate measurement data under outboard active jamming:

[0050] Based on the fact that after the jamming is released, the passive radar will not only continue to track the target but also track the jamming, while the active radar will track the equivalent centroid of the ship target and the jamming. Let the position of the target be xyz k,s =(x k,s , y k,s , z k,s ), the position of the jamming be xyz k,j2 =(x k,j2 , y k,j2 , z k,j2 ), and the position of the centroid be xyz k,c2 =(x k,c2 , y k,c2 , z k,c2 ). And the jammer works at a constant power state. Calculate the power P j of the jamming signal received by the seeker and the power P s of the target echo signal:

[0051]

[0052] Among them, P j is the jamming radiation power, P t is the seeker transmission power, G j is the jamming antenna gain, G t is the seeker antenna gain, λ is the wavelength, R1 is the distance between the jamming and the seeker, and R2 is the distance between the target and the seeker;

[0053] According to the energy relationship P j / P s =σ j / σ s , calculate the equivalent radar cross-section σ j of the jamming:

[0054]

[0055] Substitute σ j into formula <3> to calculate the position xyz k,c2 of the centroid. Let xyz k,t =xyz k,c2 , and combine formula <1> and formula <2> to calculate the measured value Z k,a2 of the active radar under outboard active jamming:

[0056]

[0057] Among them, are the measurement values of the active radar for the radial distance, azimuth angle, and pitch angle of the equivalent centroid, respectively;

[0058] Let xyz k,t = xyz k,j2 , and combine Equation <1> and Equation <2> to calculate the measurement value Z of the passive radar for the interference under the outboard active interference k,p2 :

[0059]

[0060] Among them, are the measurement values of the passive radar for the azimuth angle and pitch angle of the interference, respectively.

[0061] Step 2, filter the measurement values of the active / passive radar to obtain the state estimation quantity of the error covariance matrix P k,a of the active radar in the rectangular coordinate system and the state estimation quantity of the error covariance matrix P k,p of the passive radar in the polar coordinate system.

[0062] (2.1) Calculate the state estimation quantity of the error covariance matrix P k,a of the active radar in the rectangular coordinate system;

[0063] 2.1.1) Calculate the measurement information Z' of the active radar in the rectangular coordinate system k,a :

[0064]

[0065] Among them, are the measurement values of the radial distance, azimuth angle, and pitch angle of the active radar in the polar coordinate system, x k.a , y k,a , z k,a are the measurement values after conversion of the x, y, and z axes of the active radar in the rectangular coordinate system. Assume the state estimation quantity of the active radar in the rectangular coordinate system

[0066] 2.1.2) To avoid the problem of pseudo-acceleration caused by filtering in the polar coordinate system, it is necessary to convert the measurement data to the rectangular coordinate system for filtering, that is, calculate the prediction of the state estimation quantity of the active radar and the prediction P k|k-1,a of the error covariance matrix:

[0067]

[0068] Among them, is the state estimation quantity of the active radar at time k-1, and P k-1,a is the error covariance matrix of the active radar at time k-1, and F k,a is the state transition matrix of the active radar, and Q k,a is the state noise covariance matrix of the active radar, and [·] T represents taking the matrix transpose;

[0069] 2.1.3) Calculate the filtering innovation v of the active radar according to the measurement information Z' of the active radar in the rectangular coordinate system k,a and the predicted state estimation quantity : k,a

[0070]

[0071] Among them, H k,a is the measurement matrix of the active radar;

[0072] 2.1.4) Calculate the innovation covariance matrix S according to the predicted error covariance matrix P of the active radar k|k-1,a : k,a

[0073]

[0074] Among them, R k,a is the measurement noise covariance matrix of the active radar;

[0075] 2.1.5) Calculate the filter gain K of the active radar according to the predicted error covariance matrix P of the active radar k|k-1,a , the measurement matrix H k,a and the innovation covariance matrix S k,a : k,a

[0076]

[0077] 2.1.6) Calculate the state estimation quantity and the error covariance matrix P according to the predicted state estimation quantity of the active radar k|k-1,a , the measurement matrix H k,a , the filtering innovation v k,a and the innovation covariance matrix S k,a : and the error covariance matrix P k,a :

[0078]

[0079] (2.2) Calculate the state estimation quantity of the passive radar in the polar coordinate system and the error covariance matrix P​​​k,p :

[0080] 2.2.1) Let the state estimation variables of the passive radar in the polar coordinate system be where are the estimated variables of the azimuth angle, azimuth angular velocity, elevation angle, and elevation angular velocity of the passive radar, respectively.

[0081] 2.2.2) Calculate the prediction of the state estimation variables of the passive radar and the prediction P of the error covariance matrix k|k-1,p :

[0082]

[0083] where is the state estimation variable of the passive radar at time k - 1, and P k-1,p is the error covariance matrix of the passive radar at time k - 1, F k,p is the state transition matrix of the passive radar, and Q k,p is the state noise covariance matrix of the passive radar;

[0084] 2.2.3) According to the predicted state estimation variable of the passive radar, calculate the filtering innovation v k,p :

[0085]

[0086] where H k,p is the measurement matrix of the passive radar, and Z k,p is the measured value of the passive radar in the polar coordinate system;

[0087] 2.2.4) According to the predicted error covariance matrix P k|k-1,a of the passive radar, calculate the innovation covariance matrix S k,p :

[0088]

[0089] where R k,p is the measurement noise covariance matrix of the passive radar;

[0090] 2.2.5) According to the predicted error covariance matrix P k|k-1,p , measurement matrix H k,p and innovation covariance matrix S k,p of the passive radar, calculate the filter gain K k,p :

[0091]

[0092] 2.2.6) According to the predicted state estimation variable The predicted error covariance matrix P k|k-1,p , the measurement matrix H k,p , the filtering innovation v k,p and the innovation covariance matrix S k,p , calculate its state estimator and the error covariance matrix P k,p :

[0093]

[0094] Step 3, according to the state estimator of the active radar in the rectangular coordinate system and the error covariance P k,a , calculate the state estimator of the active radar in the polar coordinate system and the error covariance

[0095] (3.1) Calculate the state estimator of the active radar in the polar coordinate system

[0096] 3.1.1) According to the state estimator of the active radar in the rectangular coordinate system calculate the state estimators of the radial distance azimuth angle and elevation angle of the target in the polar coordinate system:

[0097]

[0098] where, are the position estimators on the x, y, and z axes in the rectangular coordinate system respectively, and (·) T represents matrix transpose;

[0099] 3.1.2) According to the state estimator of the active radar in the rectangular coordinate system calculate the state estimators of the radial velocity azimuth angular velocity and elevation angular velocity of the target in the polar coordinate system:

[0100]

[0101] where are the velocity estimators on the x, y, and z axes in the rectangular coordinate system respectively, and (·) -1 represents matrix inversion, and A(k) is the transformation matrix, expressed as:

[0102]

[0103] where, They are the measured values of the azimuth angle, elevation angle, and radial distance in the active radar polar coordinate system, respectively;

[0104] 3.1.3) Extract the angle and angular velocity information of the azimuth angle and elevation angle to obtain the state estimation quantity of the active radar in the polar coordinate system

[0105]

[0106] (3.2) Calculate the error covariance of the active radar in the polar coordinate system

[0107] 3.2.1) For the error covariance matrix P of the active radar k,a Perform the following partitioning:

[0108]

[0109] where P 11 , P 12 , P 13 , P 21 , P 22 , P 23 , P 31 , P 32 and P 33 are all 3×3 partitioned matrices;

[0110] 3.2.2) Extract the terms related to the angle and angular velocity in the partitioned matrix to form the transition covariance matrix P new :

[0111]

[0112] where and are all 3×3 partitioned matrices;

[0113] 3.2.3) Calculate the error covariance matrix R in the polar coordinate system according to the transformation matrix A(k) new :

[0114]

[0115] where and are all 3×3 partitioned matrices;

[0116] 3.2.4) Extract the quantities related to the angle and angular velocity to obtain the error covariance matrix of the active radar in the polar coordinate system

[0117]

[0118] Step 4: Perform track association on the data during the main / passive radar filtering process.

[0119] In this step, track association is used to determine the state estimators of the passive radar and the state estimators of the active radar to see if they are track estimations of the same target. The specific implementation is as follows:

[0120] (4.1) Calculate the Mahalanobis distance d between the state estimators at each moment during the filtering process k,ap :

[0121]

[0122] where is the state estimator of the active radar, the error covariance matrix of the active radar is and in the polar coordinate system is the state estimator of the passive radar, P k,p is the passive radar error covariance matrix; (·) T represents matrix transpose, (·) -1 represents matrix inversion;

[0123] (4.2) Considering connecting the track association at the current moment with that at historical moments and borrowing the idea of the sliding window method, calculate the track association statistic λ k,ap :

[0124]

[0125] where is the track association test length, T win is the size of the time window;

[0126] From the distribution of d k,ap it can be known that follows a chi-square distribution with degrees of freedom Ln, Lλ k,ap follows a chi-square distribution with degrees of freedom Ln, that is

[0127] (4.3) Use hypothesis testing to obtain the main / passive radar track association result:

[0128] 4.3.1) Denote the successful association of the active track and the passive track as event H0, and the failed association of the active track and the passive track as event H1;

[0129] 4.3.2) Let β k,ap be the discrimination threshold for track association, which satisfies: P{λ k,ap > β k,apPr{H0} = α, where α is the significance level. Usually, the value of α is 0.05, 0.01 or 0.1. In this example, α takes the value of 0.01. Pr{·} represents the probability of a certain distribution;

[0130] 4.3.3) Compare the track association statistic λ k,ap with the discrimination threshold β k,ap to determine the event to be accepted:

[0131] If λ k,ap ≤ β k,ap , then accept the event H0;

[0132] Otherwise, accept the event H1.

[0133] Step 5: Identify the type of interference according to the association result;

[0134] (5.1) Identify whether the seeker is interfered according to the association result of the active track and the passive track of the target at time k:

[0135] If the association is successful, it means that the seeker is not interfered, and proceed to Step 6;

[0136] Otherwise, it means that the seeker is interfered, and proceed to step (5.2);

[0137] (5.2) Identify the type of interference on the seeker according to the association result of the active track of the target and the passive track of the interference at time k:

[0138] If the association is successful, it means that the seeker is interfered by outboard active interference, and proceed to Step 7;

[0139] Otherwise, it means that the seeker is interfered by centroid passive interference, and proceed to Step 7;

[0140] Step 6: Calculate the state estimator based on the state estimator of the active radar and the error covariance matrix in the polar coordinate system, and the state estimator of the passive radar and the error covariance matrix P k,p in the polar coordinate system, and use it for the guidance of the seeker.

[0141]

[0141] (6.1) Let ω k,a be the fusion weight of the active track, ω k,p be the fusion weight of the passive track, ω k,p , and satisfy the following equation:

[0142]

[0143] Among them, min(·) represents finding the minimum value, and Tr[·] represents finding the trace of a matrix;

[0144] (6.2) According to the fusion weight ω of the active track k,a and the fusion weight ω of the passive track k,p , calculate the fused error covariance matrix P k and the state estimator

[0145]

[0146] Step 7, according to the state estimator of the passive radar, guide the seeker.

[0147] The following further illustrates the effect of the present invention in combination with simulation experiments:

[0148] I. Simulation experiment conditions:

[0149] The hardware platform for the simulation experiment of the present invention is: the processor is an Intel i5 7300HQ CPU with a main frequency of 2.50 GHz and a memory of 16 GB.

[0150] The software platform for the simulation experiment of the present invention is: Windows 10 Professional Edition, 64-bit operating system, MATLABR2018b.

[0151] There are two scenarios set for the simulation experiment of the present invention, among which:

[0152] Scenario 1:

[0153] Suppose the initial position of the ship target is (12000, 5000, 0) m, moving at a constant speed of 17 m / s, the course angle is (30, 0)°, and the equivalent radar cross-section is 4500 m 2 , and the process noise covariance matrix Q k is 0.

[0154] The initial position of the seeker is (0, 0, 1000) m, the speed is 500 m / s, the initial course is (0, 0)°, and the proportional navigation guidance method is used with a proportional coefficient of 2. The standard deviation of the azimuth angle and elevation angle measurement errors of the active radar is 0.2°, and the standard deviation of the ranging error is 20 m; the standard deviation of the azimuth angle and elevation angle measurement errors of the active radar is 0.8°.

[0155] The sampling interval is 0.1 s, the number of samplings is 150, and the centroid passive interference is released at the 50th sampling moment. The interference equivalent radar cross-section is 8000 m 2, the distance relative to the ship is 250 m, the angle is (180,0)°, its speed is only related to the wind speed, the wind speed is 5 m / s, and the wind direction is (130,0)°.

[0156] The operating frequency of the target radiation source is not in the same frequency band as that of the active radar, and the interference will not affect the passive radar, and the detection data of each radar has been spatially and temporally aligned.

[0157] Scenario 2:

[0158] The transmitting power of the active radar is 40000 W, and the antenna gain is 25 dB; the transmitting power of the jammer is 200 W, and the antenna gain is 2 dB.

[0159] The outboard active interference is released at the 70th sampling moment. The distance of the interference relative to the ship is 300 m, the angle is (180,0)°, the speed is only related to the wind speed, the wind speed is 5 m / s, and the wind direction is (90,0)°.

[0160] Other parameters and conditions are the same as those in Scenario 1.

[0161] II. Simulation content and its result analysis:

[0162] Simulation 1, simulate the deviation between the filtered value and the true value of the azimuth angle of the ship target by the main / passive radar under the centroid passive interference in Scenario 1. The results are as Figure 2 shown, where the solid line represents the active radar and the dashed line represents the passive radar. From Figure 2 it can be seen that the deviation of the active radar gradually increases after the 50th sampling moment, that is, the seeker may be interfered, but it is impossible to identify what kind of interference it has received.

[0163] Simulation 2, use the method of the present invention to simulate and identify the centroid passive interference in Scenario 1. The results are as Figure 3 shown. From Figure 3 it can be seen that the track correlation statistic λ k,11 of the ship target is greater than the discrimination threshold at the 55th sampling moment, that is, the track correlation fails, and there is only one correlation statistic, so there is no passive track of interference. It can be known from this that the seeker is affected by centroid passive interference. In this case, using the data of the passive radar for guidance can achieve the purpose of anti-interference.

[0164] Simulation 3, simulate the deviation between the filtered value and the true value of the azimuth angle of the ship target by the main / passive radar under the outboard active interference in Scenario 2. The results are as Figure 4 shown, where the solid line represents the active radar and the dashed line represents the passive radar. From Figure 4 it can be seen that the deviation of the active radar gradually increases after the 70th sampling moment, that is, the seeker may be interfered, but it is impossible to identify what kind of interference it has received.

[0165] Simulation 4. In scenario 2, the method of the present invention is used to simulate and identify the outboard active interference, and the results are as follows Figure 5 shown. It can be seen from Figure 5 that the track correlation statistic λ of the ship target k,11 is greater than the discrimination threshold at the 74th sampling moment, that is, the track correlation fails. At this time, the correlation statistic λ of the active track of the target and the passive track of the interference k,12 is less than the discrimination threshold. From this, it can be known that the seeker is affected by outboard active interference. In this case, the passive radar data is used for guidance to achieve the purpose of anti-interference. However, as the seeker approaches the target, the position of the equivalent centroid will gradually move away from the interference, so the correlation degree between the passive track of the interference and the active track of the equivalent centroid will decrease, and the value of λ k,12 will gradually increase.

[0166] The above experimental results verify the reliability and effectiveness of the present invention, indicating that the present invention can successfully identify and resist centroid passive interference and outboard active interference, reduce the probability of the seeker being deceived, and improve its survivability in complex environments.

Claims

1. An anti-jamming method based on data association of a combined active / passive radar seeker, characterized in that Including the following steps: (1)Obtain the measurement data of the active / passive radar from the position information of the ship target, seeker, and interference, filter it, and calculate the state estimation quantity of the active radar in the rectangular coordinate system Error covariance matrix P k,a and the state estimation quantity of the passive radar in the polar coordinate system Error covariance matrix P k,p ; (2) According to the active radar state estimator in the rectangular coordinate system and the error covariance P k,a , calculate the state estimator of the active radar in the polar coordinate system and the error covariance (3) According to the state estimators of the active radar in the polar coordinate system Error covariance matrix and the state estimators of the passive radar in the polar coordinate system Error covariance matrix P k,p , calculate the Mahalanobis distance d between the state estimators at each moment during the filtering process k,ap ; (4) According to the Mahalanobis distance d k,ap Calculate the track association statistic λ at each moment during the main / passive radar filtering process k,ap ; (5) According to the track association statistic λ k,ap , the main / passive radar track association result is obtained by using hypothesis testing; (6) Identify the type of interference according to the track association result; (6a) Identify whether the seeker is interfered according to the association result of the active track and the passive track of the target at time k: If the association is successful, it means that the seeker is not interfered, and step (7) is executed; Otherwise, it means that the seeker is interfered, and step (6b) is executed; (6b) Identify the type of interference on the seeker according to the association result of the active track of the target and the passive track of the interference at time k: If the association is successful, it means that the seeker is interfered by outboard active interference, and step (8) is executed; Otherwise, it means that the seeker is interfered by centroid passive interference, and step (8) is executed; (7) Based on the state estimators of the active radar in the polar coordinate system Error covariance matrix and the state estimators of the passive radar in the polar coordinate system Error covariance matrix P k,p , calculate the state estimator and use it for the guidance of the seeker; (8) Estimate the state of the passive radar Guide the seeker.

2. The method according to claim 1, characterized in that, Step (1) Calculate the state estimator of the active radar in the rectangular coordinate system Error covariance matrix P k,a The implementation is as follows: (1a) Calculate the measurement information Z' of the active radar in the rectangular coordinate system k,a : where r k,a , θ k.a , are the measured values of the radial distance, azimuth angle, and elevation angle in the active radar polar coordinate system, respectively, and x k.a , y k,a , z k,a are the measured values after conversion of the x, y, and z axes in the active radar rectangular coordinate system, respectively; (1b) Calculate the prediction of the active radar state estimator and the prediction P of the error covariance matrix kk-1,a : Among them, is the state estimator of the active radar at time k-1, and P k-1,a is the error covariance matrix of the active radar at time k-1, and F k,a is the state transition matrix of the active radar, and Q k,a is the state noise covariance matrix of the active radar, and [·] T denotes matrix transpose; (1c) Calculate the filtering innovation v of the active radar based on the measurement information Z' of the active radar in the rectangular coordinate system k,a and the predicted state estimator : k,a ​ Among them, H k,a is the measurement matrix of the active radar; (1d) According to the prediction error covariance matrix P of the active radar k|k-1,a Calculate the innovation covariance matrix S k,a : wherein, R k,a is the measurement noise covariance matrix of the active radar; (1e) Calculate the filter gain K of the active radar based on the predicted error covariance matrix P of the active radar k|k-1,a , the measurement matrix H k,a and the innovation covariance matrix S k,a : k,a ​ (1f)Estimate the predicted state of the active radar Prediction error covariance matrix P k|k-1,a , measurement matrix H k,a , filtering innovation v k,a and innovation covariance matrix S k,a Calculate the state estimate and error covariance matrix P k,a :

3. The method according to claim 1, wherein Step (1) Calculate the state estimator of the passive radar in the polar coordinate system Error covariance matrix P k,p , which is implemented as follows: (1g) Calculate the prediction of the passive radar state estimator and the prediction P of the error covariance matrix k|k-1,p : Among them, is the state estimation of the passive radar at time k - 1, and P k-1,p is the error covariance matrix of the passive radar at time k - 1, and F k,p is the state transition matrix of the passive radar, and Q k,p is the state noise covariance matrix of the passive radar; (1h) Estimate the predicted state of the passive radar Calculate the filtering innovation v k,p : Among them, H k,p is the measurement matrix of the passive radar, and Z k,p is the measurement value in the polar coordinate system of the passive radar; (1i) According to the prediction error covariance matrix P of the passive radar k|k-1,a Calculate the innovation covariance matrix S k,p : wherein, R k,p is the measurement noise covariance matrix of the passive radar; (1j) Calculate the filter gain K of the passive radar according to the predicted error covariance matrix P of the passive radar k|k-1,p , the measurement matrix H k,p and the innovation covariance matrix S k,p : k,p ​ (1k)Estimate the state quantity according to the predicted state of the passive radar Prediction error covariance matrix P k|k-1,p Measurement matrix H k,p Filter innovation v k,p And innovation covariance matrix S k,p Calculate its state estimate And error covariance matrix P k,p :

4. The method according to claim 1, characterized in that, In step (2), calculate the state estimator of the active radar in the rectangular coordinate system The implementation is as follows: (2a) Calculate the state estimators of the radial distance, azimuth angle, and elevation angle of the target in the polar coordinate system based on the state estimators of the active radar in the rectangular coordinate system: ​ Among them, Position estimations on the x, y, and z axes in a rectangular coordinate system respectively, (·) T Indicates finding the matrix transpose; (2b) Calculate the state estimators of the target's radial velocity, azimuth angular velocity, and pitch angular velocity in the polar coordinate system based on the state estimators of the active radar in the rectangular coordinate system. Calculate the radial velocity of the target in the polar coordinate system Azimuth angular velocity And pitch angular velocity Of the state estimators: Among them They are the velocity estimations on the x, y, and z axes in the rectangular coordinate system respectively, (·) -1 denotes the inverse of a matrix. A(k) is the transformation matrix, expressed as: where θ k.a 、 r k,a are the measured values of the azimuth angle, elevation angle, and radial distance in the polar coordinate system of the active radar, respectively; (2c)Extract the angle and angular velocity information of the azimuth and elevation angles to obtain the state estimation quantity of the active radar in the polar coordinate system 5. The method according to claim 1, characterized in that, Calculate the error covariance matrix of the active radar in the polar coordinate system in step (2). The implementation is as follows: (2d) The error covariance matrix P of the active radar k,a is partitioned as follows: Among them, P 11 , P 12 , P 13 , P 21 , P 22 , P 23 , P 31 , P 32 and P 33 are all 3×3 block matrices; 2e) Extract the terms related to angle and angular velocity in the block matrix to form the transition covariance matrix P new : wherein and are both 3×3 block matrices; 2f) Calculate the error covariance matrix R in the polar coordinate system according to the transformation matrix A(k) new : wherein and are both 3×3 block matrices; 2g) Extract the quantities related to the angle and angular velocity to obtain the error covariance matrix of the active radar in the polar coordinate system 6. The method according to claim 1, characterized in that, In step (3), based on the state estimators of the active radar in the polar coordinate system error covariance matrix and the state estimators of the passive radar in the polar coordinate system error covariance matrix P k,p , calculate the Mahalanobis distance d between the state estimators at each moment during the filtering process k,ap . The formula is as follows: Among them, (·) T represents matrix transpose, and (·) -1 represents matrix inversion.

7. The method according to claim 1, characterized in that Step (4) calculates the track association statistic λ at each moment during the filtering process of the active / passive radar according to the Mahalanobis distance d k,ap The formula is as follows for calculating the track association statistic λ at each moment during the filtering process of the active / passive radar k,ap in the following formula: Among them, is the track association test length, T win is the size of the time window.

8. According to claim 1, wherein In step (5), based on the track association statistic λ k,ap , the track association result of the active / passive radar is obtained by using hypothesis testing, and the implementation is as follows: (5a) Denote the successful association of the active track and the passive track as event H0, and the failed association of the active track and the passive track as event H1; (5b) Let β k,ap be the discrimination threshold for track association, which satisfies: P{λ k,ap > β k,ap |H0} = α, where α is the significance level and Pr{·} represents the probability of a certain distribution; (5c) Compare the track association statistic λ k,ap with the discrimination threshold β k,ap to determine the event to be accepted: If λ k,ap ≤ β k,ap , then accept event H0; otherwise, accept event H1.

9. According to claim 1, wherein In step (7), according to the state estimation quantity of the active radar in the polar coordinate system error covariance matrix and the state estimation quantity of the passive radar in the polar coordinate system error covariance matrix P k,p , calculate the state estimation quantity of the target The implementation is as follows: (7a) Let ω k,a be the fusion weight of the active track, and ω k,p be the fusion weight of the passive track ω k,p , and they satisfy the following equation: Where, min(·) represents finding the minimum value, and Tr[·] represents finding the trace of the matrix; (7b) Calculate the fused error covariance matrix P k,a and the state estimator k,p according to the fusion weights ω k of the active track and the fusion weights ω

Images