Passive networked radar target detection method based on waveform characteristics

Abstract

The present invention discloses a passive networked radar target detection method based on waveform characteristics, belonging to the field of target identification technology. The method includes obtaining a collected direct wave signal and a reflected wave signal to be detected; reading a historical database containing prior information of several historical direct waves and their emission source signals; searching the historical database for the historical direct wave with the highest similarity to the direct wave signal, and using the prior information of its emission source signal as the prior information of the direct wave signal; selecting a detector based on the prior information, inputting the direct wave signal and the reflected wave signal into the detector for detection, and outputting detection information; comparing the detection information with a preset detection threshold. If the detection information is greater than the detection threshold, a target is present; if the detection information is less than the detection threshold, a target is not present. This solution solves the problem of low target detection accuracy in low signal-to-noise ratio environments in existing networked passive radar target detection methods.

Description

Technical Field

[0001] The present invention relates to target detection technology, and in particular to a passive networked radar target detection method based on waveform characteristics. Background Art

[0002] Passive radar does not actively transmit signals, but detects the presence of a target by receiving electromagnetic signals reflected to the target. Compared with active radar, passive radar has strong stealth performance and strong anti-stealth detection capability.

[0003] Networked radar refers to the coordinated detection of multiple radars dispersed in space. By conducting multi-angle, multi-band, and multi-domain coordinated detection of spatial areas, radar networking can obtain multi-dimensional information of the target and use data fusion processing to further improve the system's target detection and recognition performance.

[0004] The schematic diagram of the current networked passive radar collaborative detection target can be referred to Figure 1 The core concept of existing networked passive radar target detectors is cross-correlation detection, which involves performing a cross-correlation (CC) between the direct wave signal and the reflected wave signal that may contain the target. The calculation results from each receiving platform are then sent to a fusion center for target detection. This method is shown in the following equation:

[0005]

[0006] in, M Represents the number of networked radar platforms, For the m The direct wave signal received by the platform, is the received reflected signal of the target to be detected; Indicates the conjugate transpose of the matrix / vector; is the set detection threshold. When it is greater than When it is less than , it means there is no target.

[0007] The above-mentioned method of cross-correlation only solves the cross-correlation between the direct wave and the reflected wave. Its detection performance is highly dependent on the accuracy of the direct wave signal extracted from the noise. This method does not fully exploit the characteristic information of the transmitted signal, making it difficult to achieve efficient target detection in special environments with low signal-to-noise ratios. In particular, when the direct wave signal is also weak, it restricts the detection capability of the passive network radar. Summary of the Invention

[0008] In view of the above-mentioned deficiencies in the prior art, the waveform feature-based passive networked radar target detection method provided by the present invention solves the problem of low target detection accuracy of the existing networked passive radar target detection method in a low signal-to-noise ratio environment.

[0009] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0010] A method for detecting a target using a passive networked radar based on waveform characteristics is provided, comprising the steps of:

[0011] S1. Obtain the collected direct wave signal and the reflected wave signal to be detected;

[0012] S2. Reading a historical database including prior information of several historical direct waves and their transmitting source signals, wherein the prior information is whether waveform characteristics of the transmitting source signal satisfy a bandwidth constraint, a similarity constraint, and / or a constant modulus constraint;

[0013] S3. Search the historical database for the historical direct wave with the highest similarity to the direct wave signal, and use the prior information of its transmitting source signal as the prior information of the direct wave signal;

[0014] S4. Select a detector based on the prior information, input the direct wave signal and the reflected wave signal into the detector for detection, and output detection information;

[0015] S5. Compare the detection information with a preset detection threshold. If the detection information is greater than the detection threshold, the target exists. If the detection information is less than the detection threshold, the target does not exist.

[0016] Furthermore, the expressions of the direct wave signal and the reflected wave signal are:

[0017] ,

[0018] in, N is the sampled signal sample; M is the number of networked radar platforms; is the direct wave signal; 、 and 1st, 2nd and M Direct wave signals received by each platform; is a matrix of dimension N×M; is the reflected wave signal; 、 and 1st, 2nd and M The reflected signal of the target to be detected is received by each platform.

[0019] Furthermore, when the waveform characteristics of the direct wave emission source signal meet the similarity constraint, the expression of the selected detector is:

[0020]

[0021] in, F The detection information output by the detector; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is:

[0022]

[0023] in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is the 2-norm of the vector.

[0024] Furthermore, when the waveform characteristics of the direct wave emission source signal meet the similarity constraint and bandwidth constraint, the expression of the selected detector is:

[0025]

[0026] in, F The detection information output by the detector; is the projection matrix on the subspace of the bandwidth matrix column; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is:

[0027]

[0028] in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true,i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is the 2-norm of the vector.

[0029] Furthermore, the optimization problem is solved using the nearest vector problem method:

[0030]

[0031] Among them, max is the maximum value; and Respectively k +1 and k The emission waveform at the iteration; the iteration termination condition is , is the termination threshold.

[0032] Furthermore, when the waveform characteristics of the direct wave emission source signal meet the similarity constraint and bandwidth constraint, the expression of the selected detector is:

[0033]

[0034]

[0035] in, F The detection information output by the detector; and They represent the estimated values ​​under the assumption that the detection target does not exist and the assumption that the detection target exists, respectively. i is a variable, equal to 0 or 1; is the vector 2 norm; For the k +1 emission waveform at iteration time; and Respectively k The attenuation of the direct wave signal during the propagation process and the attenuation of the echo signal after being reflected by the target at the iteration; is the conjugate operation.

[0036] Further, the k Emission waveform at +1 iteration The methods for obtaining include:

[0037] A1. Initialize the number of iterations k =0 and initial emission wave waveform , and set the termination threshold ;

[0038] A2. Judgment Is it true? If so, output 、 and Otherwise, go to step A3. For the k The emission waveform at the iteration;

[0039] A3. Calculation and The expressions are:

[0040]

[0041] in,(·) H is the conjugate transpose;

[0042] A4. Calculate the k The optimal solution at +1 iteration :

[0043]

[0044] Among them, min is the minimum value; is the independent variable of the optimization problem; is a linear frequency modulation signal;

[0045] A4. According to k The optimal solution at +1 iteration is calculated k Emission waveform at +1 iteration :

[0046]

[0047]

[0048] in, For Maximum likelihood estimate of the modulus; is the projection matrix on the subspace of the bandwidth matrix column; Re(·) is the real part operation.

[0049] Furthermore, when the waveform characteristics of the direct wave emission source signal satisfy the constant modulus constraint and the similarity constraint, the expression of the selected detector is:

[0050]

[0051] in, F The detection information output by the detector; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is:

[0052]

[0053] in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is a constant modulus constraint.

[0054] Furthermore, calculation and The expression formula is:

[0055]

[0056] Among them, Y r 、Y r H are the direct wave signal and the conjugate transpose of the direct wave signal, Y s 、Y s H are the reflected wave signal and the conjugate transpose of the reflected wave signal respectively.

[0057] Furthermore, calculation The expression is:

[0058]

[0059] in, is the bandwidth matrix; (·) -1 Inverse operation for the matrix; for The conjugate transpose of .

[0060] The beneficial effects of the present invention are as follows: when performing target detection, this scheme can accurately obtain the prior information of the emission source of the transmitted wave signal through similarity queries in the historical database, and accurately select the detector based on the prior information, so as to ensure that the selected detector can better eliminate the amplitude deviation value caused by noise, etc., thereby improving the detector performance and thus improving the detection capability of the passive network radar.

[0061] When the waveform characteristics of the direct wave's transmitting source signal meet the similarity constraint, the selected detector can make full use of the prior waveform information of the transmitting source signal, that is, constrain the correlation between the received waveform information and historical information; the detector not only utilizes the information of the current received signal, but can also make full use of historical prior information, eliminating the situation in which the signal deviates from the true value interval due to noise in the detector, thereby improving the overall detection capability of the detector.

[0062] When the waveform characteristics of the direct wave's source signal meet similarity constraints and bandwidth constraints, the two selected detectors not only utilize the prior waveform information of the source signal, but also introduce waveform bandwidth information; this bandwidth information can eliminate out-of-band noise components and constrain the waveform characteristics within the specified filter range, which can significantly improve the detector's detection capability.

[0063] When the waveform characteristics of the direct wave's transmitter signal satisfy the constant modulus constraint and similarity constraint, the selected detector introduces a constant modulus information constraint on the waveform based on the consideration of the prior waveform information. The constant modulus constraint characterizes that the amplitude of the transmitter signal is constant, with only the phase changing. This constraint can keep the signal in the detector at a constant amplitude, eliminating amplitude deviations caused by noise in the detector and improving the detector performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 Schematic diagram of networked passive radars collaboratively detecting targets.

[0065] Figure 2 The figure is a flow chart of the passive network radar target detection method based on waveform characteristics.

[0066] Figure 3 This is a simulation comparison diagram of target detection performed by five detectors in the embodiment. DETAILED DESCRIPTION

[0067] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0068] refer to Figure 2 , Figure 3 FIG. 4 shows a flow chart of a passive network radar target detection method based on waveform characteristics; FIG. Figure 2 As shown, the method S includes steps S1 to S5.

[0069] In step S1, the collected direct wave signal and the reflected wave signal to be detected are obtained; specifically, the expressions of the direct wave signal and the reflected wave signal are respectively:

[0070] ,

[0071] in, N is the sampled signal sample; M is the number of networked radar platforms; is the direct wave signal; 、 and 1st, 2nd and M Direct wave signals received by each platform; is a matrix of dimension N×M; is the reflected wave signal; 、 and 1st, 2nd and M The reflected signal of the target to be detected is received by each platform.

[0072] In step S2, a historical database including prior information of several historical direct waves and their transmitting source signals is read, wherein the prior information is whether the waveform characteristics of the transmitting source signal satisfy the bandwidth constraint, similarity constraint and / or constant modulus constraint.

[0073] In step S3, the historical direct wave with the highest similarity to the direct wave signal is searched in the historical database, and the prior information of its transmitting source signal is used as the prior information of the direct wave signal;

[0074] In step S4, a detector is selected according to the prior information, and the direct wave signal and the reflected wave signal are input into the detector for detection, and detection information is output;

[0075] In step S5, the detection information is compared with a preset detection threshold. If the detection information is greater than the detection threshold, the target exists; if the detection information is less than the detection threshold, the target does not exist.

[0076] In one embodiment of the present invention, when the waveform characteristics of the direct wave emission source signal meet the similarity constraint, the expression of the selected detector is:

[0077]

[0078] in, F The detection information output by the detector; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is:

[0079]

[0080] in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is the 2-norm of the vector.

[0081] In one embodiment of the present invention, when the waveform characteristics of the direct wave emission source signal meet the similarity constraint and the bandwidth constraint, the expression of the selected detector is:

[0082]

[0083] in, F The detection information output by the detector; is the projection matrix on the subspace of the bandwidth matrix column; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is:

[0084]

[0085] in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is the 2-norm of the vector.

[0086] During implementation, this solution preferably uses the nearest vector problem method to solve the above optimization problem:

[0087]

[0088] Among them, max is the maximum value; and Respectively k +1 and k The emission waveform at the iteration; the iteration termination condition is , is the termination threshold; the optimization process is an iterative process, k represents the number of iterations, and the final waveform information can be obtained until the optimization problem converges.

[0089] In one embodiment of the present invention, when the waveform characteristics of the direct wave emission source signal meet the similarity constraint and the bandwidth constraint, the expression of the selected detector is:

[0090]

[0091]

[0092] in, F The detection information output by the detector; and They represent the estimated values ​​under the assumption that the detection target does not exist and the assumption that the detection target exists, respectively. i is a variable, equal to 0 or 1; is the vector 2 norm; For the k +1 emission waveform at iteration time; and Respectively k The attenuation of the direct wave signal during the propagation process and the attenuation of the echo signal after being reflected by the target at the iteration; is the conjugate operation.

[0093] When implementing, this plan is preferred k Emission waveform at +1 iteration The methods for obtaining include:

[0094] A1. Initialize the number of iterations k =0 and initial emission wave waveform , and set the termination threshold ;

[0095] A2. Judgment Is it true? If so, output 、 and Otherwise, go to step A3. For the k The emission waveform at the iteration;

[0096] A3. Calculation and The expressions are:

[0097]

[0098] in,(·) H is the conjugate transpose;

[0099] A4. Calculate the k The optimal solution at +1 iteration :

[0100]

[0101] Among them, min is the minimum value; is the independent variable of the optimization problem; is a linear frequency modulation signal;

[0102] A4. According to k The optimal solution at +1 iteration is calculated k Emission waveform at +1 iteration :

[0103]

[0104]

[0105] in, For Maximum likelihood estimate of the modulus; is the projection matrix on the subspace of the bandwidth matrix column; Re(·) is the real part operation.

[0106] In one embodiment of the present invention, when the waveform characteristics of the direct wave emission source signal satisfy the constant modulus constraint and the similarity constraint, the expression of the selected detector is:

[0107]

[0108] in, F The detection information output by the detector; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is:

[0109]

[0110] in,s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is a constant modulus constraint.

[0111] When implemented, this solution preferably calculates and The expression formula is:

[0112]

[0113] Among them, Y r 、Y r H are the direct wave signal and the conjugate transpose of the direct wave signal, Y s 、Y s H are the reflected wave signal and the conjugate transpose of the reflected wave signal respectively.

[0114] calculate The expression is:

[0115]

[0116] in, is the bandwidth matrix; (·) -1 Inverse operation for the matrix; for The conjugate transpose of .

[0117] The following simulation demonstrates how the passive network radar target detection method of this solution can improve target detection performance under low signal-to-noise ratio conditions:

[0118] Three distributed passive network radar receiving platforms are used in the simulation, each of which has the ability to receive direct waves and reflected waves. The number of sampling points is set to 64, the signal bandwidth is 500MHz, and the signal source is a constant modulus signal. The detection probability under different input signal-to-noise ratio conditions is simulated, and the false alarm probability is used as , 1000 Monte Carlo experiments were conducted at each point; historical data similarity waveform source was used.

[0119] Among them, the CC detector is a detector that does not consider any constraints, and its expression is:

[0120] .

[0121] The SP detector is selected when the waveform characteristics of the direct wave emission source signal meet the similarity constraint; the SPBC detector and the SCBC detector are both selected when the waveform characteristics of the direct wave emission source signal meet the similarity constraint and bandwidth constraint; the SPC detector is selected when the waveform characteristics of the direct wave emission source signal meet the constant modulus constraint and similarity constraint.

[0122] The above five detectors are used to simulate the passive network radar target detection method. The simulation results are as follows Figure 3 All results in the figure are from experiments conducted in the same scenario. Compared to the CC detector without any constraints, the SP detector significantly improves detection probability. For methods that use both similarity and bandwidth constraints, the SPBC and SCBC detectors perform comparable and better than the SP detector that only considers similarity constraints. The SPC detector, which considers both constant modulus and similarity constraints, achieves the best performance.

[0123] In summary, the prior information of the waveform can make the performance of the designed detector better, and the more prior information of the waveform is obtained, the better the performance of the detector will be.

Claims

1. A passive network radar target detection method based on waveform characteristics, characterized in that: Including steps: S1. Obtain the collected direct wave signal and the reflected wave signal to be detected; S2. Reading a historical database including prior information of several historical direct waves and their transmitting source signals, wherein the prior information is whether waveform characteristics of the transmitting source signal satisfy a bandwidth constraint, a similarity constraint, and / or a constant modulus constraint; S3. Search the historical database for the historical direct wave with the highest similarity to the direct wave signal, and use the prior information of its transmitting source signal as the prior information of the direct wave signal; S4. Select a detector based on the prior information, input the direct wave signal and the reflected wave signal into the detector for detection, and output detection information; S5. Compare the detection information with a preset detection threshold. If the detection information is greater than the detection threshold, the target exists. If the detection information is less than the detection threshold, the target does not exist.

2. The passive network radar target detection method according to claim 1, characterized in that: The expressions of direct wave signal and reflected wave signal are: ， in, N is the sampled signal sample; M is the number of networked radar platforms; is the direct wave signal; 、 and 1st, 2nd and M Direct wave signals received by each platform; is a matrix of dimension N×M; is the reflected wave signal; 、 and 1st, 2nd and M The reflected signal of the target to be detected is received by each platform.

3. The passive network radar target detection method according to claim 2, characterized in that: When the waveform characteristics of the direct wave source signal meet the similarity constraint, the expression of the selected detector is: in, F The detection information output by the detector; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is: in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is the 2-norm of the vector.

4. The passive network radar target detection method according to claim 2, characterized in that: When the waveform characteristics of the direct wave emission source signal meet the similarity constraint and bandwidth constraint, the expression of the selected detector is: in, F The detection information output by the detector; is the projection matrix on the subspace of the bandwidth matrix column; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is: in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is the 2-norm of the vector.

5. The passive network radar target detection method according to claim 4, characterized in that: The optimization problem is solved using the nearest vector problem method: Among them, max is the maximum value; and Respectively k +1 and k The emission waveform at the iteration; the iteration termination condition is , is the termination threshold.

6. The passive network radar target detection method according to claim 2, characterized in that: When the waveform characteristics of the direct wave emission source signal meet the similarity constraint and bandwidth constraint, the expression of the selected detector is: in, F The detection information output by the detector; and They represent the estimated values ​​under the assumption that the detection target does not exist and the assumption that the detection target exists, respectively. i is a variable, equal to 0 or 1; is the vector 2 norm; For the k +1 emission waveform at iteration time; and Respectively k The attenuation of the direct wave signal during the propagation process and the attenuation of the echo signal after being reflected by the target at the iteration; is the conjugate operation; is the projection matrix onto the subspace of the bandwidth matrix columns.

7. The passive network radar target detection method according to claim 6, characterized in that: No. k Emission waveform at +1 iteration The methods for obtaining include: A1. Initialize the number of iterations k =0 and initial emission wave waveform , and set the termination threshold ; A2. Judgment Is it true? If so, output 、 and Otherwise, go to step A3. For the k The emission waveform at the iteration; A3. Calculation and The expressions are: in,(·) H is the conjugate transpose; A4. Calculate the k The optimal solution at +1 iteration : Among them, min is the minimum value; is the independent variable of the optimization problem; is a linear frequency modulation signal; A5. According to k The optimal solution at +1 iteration is calculated k Emission waveform at +1 iteration : in, For Maximum likelihood estimate of the modulus; is the projection matrix on the subspace of the bandwidth matrix column; Re(·) is the real part operation; and are the attenuation of the amplitude of the direct wave signal during the propagation process and the attenuation of the amplitude of the echo signal after being reflected by the target at the kth iteration.

8. The passive network radar target detection method according to claim 2, characterized in that: When the waveform characteristics of the direct wave source signal satisfy the constant modulus constraint and similarity constraint, the expression of the selected detector is: in, F The detection information output by the detector; is the direct wave signal matrix; is the superposition value of the reflected wave signal matrix and the direct wave signal matrix after being reflected by the target; (·) H is the conjugate transpose; and They are and The solution to the corresponding optimization problem is: in, s is the independent variable of the optimization problem; i is a variable. When the hypothesis is not detected to be true, i =0, when the hypothesis detects that the target is established, i =1; is the similarity coefficient; is a linear frequency modulation signal; min is the minimum value; is a constant modulus constraint.

9. The passive network radar target detection method according to any one of claims 3 to 5, characterized in that: calculate and The expression formula is: Among them, Y r 、Y r H are the direct wave signal and the conjugate transpose of the direct wave signal, Y s 、Y s H are the reflected wave signal and the conjugate transpose of the reflected wave signal respectively.

10. The passive network radar target detection method according to any one of claims 4 to 7, characterized in that: calculate The expression is: in, is the bandwidth matrix; (·) -1 Inverse operation for the matrix; for The conjugate transpose of .

Images