Field programmable gate array (FPGA)-based broadband high-precision quadratic correlation interferometer direction finding method

Abstract

The invention belongs to the technical field of interference direction finding, and relates to a broadband high-precision quadratic correlation interferometer direction finding method based on an FPGA. The method comprises the following steps: carrying out first phase correlation calculation by adopting an FPGA parallel calculation mode; performing second phase correlation calculation; performing quadric surface fitting on a curved surface formed by extending in the horizontal direction and the pitching direction; traversing all sample angles, calculating a matrix coefficient required by quadric surface fitting and an inverse matrix of a matrix, and storing a result in a BRAM (Block Random Access Memory) of the FPGA; solving a quadric surface coefficient to obtain a surface equation; and calculating a final horizontal angle and a final pitch angle according to the extreme points of the curved surface equation. According to the invention, quadratic fitting of the curved surface is realized in the FPGA based on a result storage mode, so that the direction finding precision of the correlation interferometer is improved; the phase correlation calculation and resource overhead of the direction finding of the primary correlation interferometer are obviously reduced, and the bandwidth performance and the real-time performance of direction finding are ensured while the direction finding precision is effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of interference direction finding, and specifically relates to a wideband high-precision quadratic correlation interferometer direction finding method based on FPGA. Background Technique

[0002] The interferometer direction finding method is widely used in fields such as communication, radio monitoring, and radio astronomy. According to the phase relationship generated by the projection distance difference of electromagnetic waves arriving at different array elements of the antenna array, the direction of arrival of radio signals is determined. Since the signal azimuth corresponds one-to-one with the phase distribution of the array element signals, the direction of arrival information can be obtained. This method has the advantages of high direction finding accuracy, high sensitivity, and good real-time performance. However, in actual applications, affected by factors such as antenna manufacturing technology, installation location, and working environment, the consistency of antenna channels is poor, and the phase and amplitude of the signals received by each array element will be distorted and distorted, resulting in an increase in direction finding error.

[0003] By comparing the similarity between the incident wave and the phase distributions of incoming waves with known azimuths and frequencies, the direction of arrival of the incoming wave can be determined, which can offset the influence of the inconsistency of the antenna array elements to a certain extent.

[0004] The related interferometer direction finding method still has limitations, and the sample interval has a great influence on the direction finding accuracy. A small sample interval can improve the accuracy, but it will cause a huge increase in the amount of calculation, which is difficult to implement on an FPGA (Field-Programmable Gate Array) platform, easily exhausts the chip resources and increases the power consumption, affecting the stability and reliability of the system. Therefore, how to balance the direction finding accuracy with the amount of calculation and resource overhead is the key problem faced by this technology at present. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a wideband high-precision quadratic correlation interferometer direction finding method based on FPGA, including: Adopting the parallel computing method of FPGA, performing the first phase correlation calculation on the signal by stepping the first degree in the horizontal direction and the second degree in the pitch direction, and obtaining the first horizontal direction information and the first pitch direction information corresponding to the maximum value of the phase correlation; According to the result of the first phase correlation calculation, taking the horizontal direction information and the pitch direction information as the center, stepping in the horizontal direction with the first degree range and the third degree, and stepping in the pitch direction with the second degree range and the fourth degree, performing the second phase correlation calculation, and obtaining the second horizontal direction information and the second pitch direction information corresponding to the maximum value of the phase correlation; Selecting to perform quadratic surface fitting on the surface formed by extending in the horizontal direction and the pitch direction with the second horizontal direction information and the second pitch direction information as the center; Traverse all sample angles, calculate the matrix coefficients and the inverse matrix required for quadratic surface fitting, and store the results in the BRAM of the FPGA; Using the pre-stored matrix data, combined with the relevant values of all points, solve the quadratic surface coefficients through matrix operations to obtain the surface equation; According to the extreme points of the surface equation, calculate the final horizontal angle and pitch angle as the calculation result of the incoming wave direction.

[0006] On the basis of the above technical solution, the present invention can also be improved as follows.

[0007] Further, after the antenna array elements receive the incoming wave signals, perform synchronous acquisition and channelization processing, and calculate and record the amplitude information and phase information of the signals.

[0008] Further, the first degree is greater than the third degree, and the second degree is greater than the fourth degree.

[0009] Further, when performing the first phase correlation calculation, the sample steps 10 degrees in the horizontal direction and 8 degrees in the pitch direction.

[0010] Further, when performing the second phase correlation calculation, the sample takes the first degree range in the horizontal direction, steps 5 degrees, and takes the second degree range in the pitch direction, steps 2 degrees; let the horizontal direction information be , and the pitch direction information be ; the first degree range is , and the second degree range is .

[0011] Further, when performing channelization processing, preset the sample parameters as a horizontal angle step of 5 degrees and a pitch angle step of 2 degrees; preset the number of baselines selected by the direction finding system.

[0012] Further, the horizontal angle steps 5 degrees, a total of 72 azimuths, and the pitch angle steps 2 degrees, a total of 44 azimuths.

[0013] Further, let the horizontal angle be , and the pitch angle be , the number of baselines selected by the direction finding system is , is the selected root baseline at the horizontal angle of and the pitch angle of direction sample phase and signal phase correlation value, is the selected root baseline at the horizontal angle of and the pitch angle of direction phase difference, is the root baseline electromagnetic wave phase difference, then: 。

[0014] Furthermore, for the selected radical line, the sample phase at a horizontal angle of and a pitch angle of is related to the signal phase, and the angle corresponding to the maximum value of the correlation value is the direction of arrival of the electromagnetic wave with the highest probability.

[0015] Furthermore, the quadratic surface is a 3×3 surface with a total of 9 points. Let the horizontal angle be and the pitch angle be . The number of baselines selected by the direction finding system is , is the abscissa of the quadratic surface, is the ordinate of the quadratic surface, represents the function of the quadratic surface, represents the coefficients of the quadratic surface equation, where , then the quadratic surface equation is expressed as ; Let the function value of the quadratic surface be , represents the correlation value of the -th point, represents the horizontal angle of the -th point, represents the pitch angle of the -th point, is the correlation value between the sample phase and the signal phase at the selected horizontal angle of and the pitch angle of , ; Let the coefficients of the quadratic surface be , , . Let: ; ; Let the matrix be the matrix calculation result of . Traverse all matrices formed by the relevant sample angles, and calculate . Store all corresponding to the matrix matrix calculation results in the BRAM of the FPGA.

[0016] The beneficial effects of the present invention are as follows: The present invention realizes the quadratic fitting of the curved surface in the FPGA by means of result storage, so as to improve the direction-finding accuracy of the relevant interferometer; through the method of quadratic correlation, without sacrificing accuracy, the phase correlation calculation and resource overhead of the first-order correlation interferometer direction-finding are significantly reduced, meeting the application scenario of wideband real-time direction-finding processing, and ensuring the bandwidth performance and real-time performance of direction-finding while effectively improving the direction-finding accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 FIG. is a schematic diagram of the method for wideband high-precision quadratic correlation interferometer direction-finding based on FPGA provided in Embodiment 1 of the present invention.

[0018] Figure 2 FIG. is a flowchart of the direction-finding data stream of the relevant interferometer; Figure 3 FIG. is a schematic diagram of the curved surface formed by selecting the vicinity of the maximum correlation value. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations.

[0020] As an embodiment, as shown in the attached Figure 1 figures, to solve the above technical problems, the present embodiment provides a method for wideband high-precision quadratic correlation interferometer direction-finding based on FPGA, including: Adopting the parallel computing method of FPGA, performing the first phase correlation calculation on the signal by stepping the first degree in the horizontal direction and the second degree in the pitch direction, and obtaining the first horizontal direction information and the first pitch direction information corresponding to the maximum phase correlation value; According to the results of the first phase correlation calculation, taking the horizontal direction information and the pitch direction information as the center, stepping in the horizontal direction within the range of the first degree and the third degree and stepping in the pitch direction within the range of the second degree and the fourth degree, performing the second phase correlation calculation, and obtaining the second horizontal direction information and the second pitch direction information corresponding to the maximum phase correlation value; Selecting the second horizontal direction information and the second pitch direction information as the center, and performing quadratic surface fitting on the curved surface formed by extending in the horizontal direction and the pitch direction; Traversing all sample angles, calculating the matrix coefficients and the inverse matrix required for quadratic surface fitting, and storing the results in the BRAM (Block RAM) of the FPGA. Using the pre-stored matrix data, combining the relevant values of all points, solving for the quadratic surface coefficients through matrix operations, and obtaining the surface equation; According to the extreme points of the surface equation, calculate the final horizontal angle and pitch angle as the calculation result of the incoming wave direction.

[0021] The present invention realizes the quadratic fitting of the surface in the FPGA by adopting a result storage-based method, achieving the purpose of improving the direction finding accuracy of the relevant interferometer; through the method of quadratic correlation, without sacrificing accuracy, significantly reducing the phase correlation calculation and resource overhead of the first-order correlation interferometer direction finding, and meeting the application scenario of broadband real-time direction finding processing.

[0022] Optionally, after the antenna array elements receive the incoming wave signals, synchronous acquisition and channelization processing are performed, and the amplitude information and phase information of the signals are calculated and recorded.

[0023] As shown in the Figure 2 attachment, the number of baselines selected by the direction finding system is 5, including antenna 1, antenna 2, antenna 3, antenna 4, and antenna 5. After the signals are collected by 5 ADC (Analog-to-Digital Converter) collectors, broadband DDC (Digital Down Conversion) processing is performed, the amplitude information and phase information are extracted, the first phase correlation calculation, the second phase correlation calculation, and quadratic surface fitting are performed, and finally the optimal direction finding result is obtained.

[0024] Optionally, the first degree is greater than the third degree, and the second degree is greater than the fourth degree.

[0025] Optionally, when performing the first phase correlation calculation, the sample steps 10 degrees in the horizontal direction and 8 degrees in the pitch direction.

[0026] Optionally, when performing the second phase correlation calculation, the sample takes the first degree range in the horizontal direction, steps 5 degrees, and takes the second degree range in the pitch direction, steps 2 degrees; let the horizontal direction information be and the pitch direction information be ; the first degree range is and the second degree range is .

[0027] Optionally, when performing channelization processing, the preset sample parameters are a horizontal angle step of 5 degrees and a pitch angle step of 2 degrees; the preset number of baselines selected by the direction finding system.

[0028] Optionally, the horizontal angle steps 5 degrees, with a total of 72 azimuths, and the pitch angle steps 2 degrees, with a total of 44 azimuths.

[0029] As a comparison, the number of correlation calculations in the traditional method: Let is the number of baselines selected for the direction finding system, , is the number of horizontal samples, , is the number of pitch samples, , then: ; In the secondary correlation method of the present invention, the number of correlation calculations: Let be the number of horizontal samples for the second correlation, , be the number of pitch samples for the second correlation, , then: . Through two - stage correlation calculations, while ensuring the angular accuracy of the correlation samples and without loss of the correlation calculation accuracy, the phase - correlation calculation overhead of the single - stage correlation interferometer direction finding is significantly reduced.

[0030] Optionally, let the horizontal angle be , the pitch angle be , the number of baselines selected for the direction finding system be , be the correlation value between the sample phase and the signal phase of the th baseline at the horizontal angle of and the pitch angle of , be the phase difference of the th baseline at the horizontal angle of and the pitch angle of , be the electromagnetic wave phase difference of the th baseline, then: .

[0031] Optionally, the angle corresponding to the maximum value of the correlation value between the sample phase and the signal phase of the th baseline at the horizontal angle of and the pitch angle of is the direction of the electromagnetic wave with the highest probability of arrival.

[0032] After determining the second - level horizontal direction information and the second - level pitch direction information , to improve the direction - finding accuracy of the correlation interferometer direction - finding system, a quadratic - surface fitting of the surface formed by the correlation result is performed near . Select a point extended in each of the pitch direction and the horizontal direction with as the center, and a surface composed of 3 * 3 points is formed, as shown in Appendix Figure 3 , , , , , , , , and constitute a 3×3 surface.

[0033] Optionally, the quadratic surface is a 3×3 surface with a total of 9 points. Let the horizontal angle be , the pitch angle be , the number of selected baselines of the direction finding system be , is the abscissa of the quadratic surface, is the ordinate of the quadratic surface, represents the function of the quadratic surface, represents the coefficients of the quadratic surface equation, where , then the quadratic surface equation is expressed as ; Let the function value of the quadratic surface be , represents the relevant value of the th point, represents the horizontal angle of the th point, represents the pitch angle of the th point, is the correlation value between the sample phase and the signal phase in the direction of the selected horizontal angle and the pitch angle ; ; Let the coefficients of the quadratic surface be , , , let: ; ; Let the matrix calculation result of the matrix be . Traverse all the matrices formed by the relevant sample angles, and calculate . Store all the corresponding matrix matrix calculation results in the BRAM of the FPGA.

[0034] The direction finding result is: .

[0035] Since the above calculation involves matrix calculation, and the matrix inversion process is complex and difficult to implement based on FPGA, the present invention indirectly completes the calculation of the surface coefficients by a storage method. Since the matrix The coefficient is only related to the corresponding angles of the relevant samples. Therefore, the method adopted is to traverse all the matrices formed by the angles of the relevant samples , and calculate . For all corresponding matrices of the matrix calculation results are stored in the BRAM of the FPGA, and the storage space occupies 42KB.

[0036] The present invention can significantly improve the direction-finding accuracy of the relevant interferometer, is no longer limited by the selection of the sample angle step, and at the same time, the direction-finding calculation cost is significantly reduced compared with the traditional algorithm, ensuring the bandwidth performance and real-time performance of direction-finding while effectively improving the direction-finding accuracy.

[0037] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A broadband high-precision two-stage correlation interferometer direction finding method based on FPGA, characterized in that Including: Adopting the parallel computing method of FPGA, performing the first phase correlation calculation on the signal by stepping the first degree in the horizontal direction and the second degree in the pitch direction, and obtaining the first horizontal direction information and the first pitch direction information corresponding to the maximum value of the phase correlation; According to the result of the first phase correlation calculation, taking the horizontal direction information and the pitch direction information as the center, stepping in the horizontal direction with the first degree range and the third degree, and stepping in the pitch direction with the second degree range and the fourth degree, performing the second phase correlation calculation, and obtaining the second horizontal direction information and the second pitch direction information corresponding to the maximum value of the phase correlation; Selecting to perform quadratic surface fitting on the surface formed by extending in the horizontal direction and the pitch direction with the second horizontal direction information and the second pitch direction information as the center; Traversing all sample angles, calculating the matrix coefficients and the inverse matrix required for quadratic surface fitting, and storing the results in the BRAM of the FPGA; Using the pre-stored matrix data, combining the correlation values of all points, and solving the quadratic surface coefficients through matrix operations to obtain the surface equation; Calculating the final horizontal angle and pitch angle according to the extreme points of the surface equation as the calculation result of the incoming wave direction.

2. The method for direction finding of a broadband high-precision quadratic correlation interferometer based on FPGA according to claim 1, wherein After the antenna array elements receive the incoming wave signal, perform synchronous acquisition and channelization processing, and calculate and record the amplitude information and phase information of the signal.

3. The method for direction finding of a wideband high-precision quadratic correlation interferometer based on FPGA according to claim 1, wherein, The first degree is greater than the third degree, and the second degree is greater than the fourth degree.

4. The method for direction finding of a wideband high-precision secondary correlation interferometer based on FPGA according to claim 1, wherein When performing the first phase correlation calculation, the sample steps 10 degrees in the horizontal direction and 8 degrees in the pitch direction.

5. The method for direction finding of a wideband high-precision quadratic correlation interferometer based on FPGA according to claim 1, characterized in that When performing the second phase correlation calculation, the sample takes the first degree range in the horizontal direction with a step of 5 degrees and the second degree range in the pitch direction with a step of 2 degrees; let the horizontal direction information be , and the pitch direction information be ; the first degree range is , and the second degree range is .

6. The method for direction finding of a broadband high-precision quadratic correlation interferometer based on FPGA according to claim 1, characterized in that When performing channelization processing, preset the sample parameters as the horizontal angle stepping 5 degrees and the pitch angle stepping 2 degrees; preset the number of baselines selected by the direction finding system.

7. The method for direction finding of a broadband high-precision quadratic correlation interferometer based on FPGA according to claim 6, wherein The horizontal angle steps 5 degrees, with a total of 72 azimuths, and the pitch angle steps 2 degrees, with a total of 44 azimuths.

8. The method for direction finding of a broadband high-precision quadratic correlation interferometer based on FPGA according to claim 1, wherein it is assumed that The horizontal angle is , the pitch angle is , the number of baselines selected by the direction finding system is , is the correlation value between the sample phase and the signal phase of the th selected basic baseline at a horizontal angle of and a pitch angle of . is the phase difference of the th selected basic baseline at a horizontal angle of and a pitch angle of . is the electromagnetic wave phase difference of the th basic baseline. Then: 。 9. The method for direction finding of a wideband high-precision quadratic correlation interferometer based on FPGA according to claim 8, wherein The selected root line has a horizontal angle of and a pitch angle of The angle corresponding to the maximum value of the sample phase correlation value with the signal phase in the direction of is the direction of arrival of the electromagnetic wave with the highest probability.

10. The method for direction finding of a wideband high-precision quadratic correlation interferometer based on FPGA according to claim 1, wherein The quadric surface is a 3*3 surface with a total of 9 points. Let the horizontal angle be , the pitch angle be , the number of baselines selected by the direction finding system be , be the abscissa of the quadric surface, be the ordinate of the quadric surface, represent the function of the quadric surface, represent the coefficients of the quadric surface equation, where , then the quadric surface equation is expressed as ; Let the function value of the quadric surface be , represent the correlation value of the th point, represent the horizontal angle of the th point, represent the pitch angle of the th point, be the correlation value of the sample phase and the signal phase in the direction where the selected horizontal angle is and the pitch angle is , ; Let the coefficients of the quadric surface be , , Let: ； ； Let the matrix The matrix calculation result of is, traverse all the matrices formed by the relevant sample angles and calculate For all the corresponding matrix store the matrix calculation result in the BRAM of the FPGA.

Images