Phased array tracking method based on four-dimensional joint weighted dynamic pre-switching

Abstract

The invention relates to a phased array tracking method based on four-dimensional joint weighted dynamic pre-switching, and belongs to the technical field of phased array communication. The method comprises the following steps: updating space-time correlation parameters to a target historical motion track database, and generating a target state prediction value based on a UKF prediction method; starting a four-dimensional joint gradient weighting program, carrying out linear phase interpolation of a time delay dimension, and generating a control signal with a continuous phase; solving a mutual coupling matrix, establishing a cross-array-plane phase mapping model based on cubic spline interpolation, and compensating a residual phase error; according to the real-time radar cross section, the number of effective array elements participating in beam forming is dynamically adjusted, a final beam pointing signal is generated, and stable relay tracking of the current period is completed. The method solves the problems of phase jump and signal attenuation during multi-array-plane switching, and is suitable for continuous tracking in a high-speed moving scene of the unmanned aerial vehicle.

Description

Technical Field

[0001] The present invention belongs to the technical field of phased array communication, and specifically relates to a phased array tracking method based on four-dimensional joint weighted dynamic pre-switching, which is applicable to the stable relay tracking of vehicle-mounted multi-band phased array antennas. Background Art

[0002] Traditional phased array multi-panel switching technologies mostly adopt single-dimensional compensation strategies. For example, beam transition is achieved only through linear phase interpolation in the azimuth or elevation dimension. Such methods have the following defects:

[0003] (1) Poor phase continuity: Single-dimensional compensation cannot suppress polarization mismatch and delay jump, resulting in a high peak value of phase error during panel switching and a decrease in beam pointing accuracy;

[0004] (2) Insufficient dynamic response: The pre-switching algorithm has a large prediction error for non-linear moving targets, and the switching delay often exceeds 2 beam periods, causing tracking interruption;

[0005] (3) Limited calibration accuracy: Compensation depends on the signal of the ground calibration station and does not combine the synchronous satellite beacon, and the root mean square value of the residual phase error is relatively high, making it difficult to meet the high-precision tracking requirements;

[0006] (4) Low energy efficiency ratio: The fixed array element activation ratio strategy wastes significant power in low RCS scenarios;

[0007] In summary, the prior art has not realized the joint gradual change weighting of four-dimensional parameters of azimuth, elevation, polarization, and delay, nor does it have an adaptive cooperation mechanism based on the target motion characteristics and RCS response. Therefore, there is an urgent need for a multi-panel stable tracking scheme that can simultaneously solve phase jump, prediction lag, calibration error, and energy consumption optimization. Summary of the Invention

[0008] In view of this, the present invention provides a phased array tracking method based on four-dimensional joint weighted dynamic pre-switching. The present invention can achieve stable relay tracking of a four-sided phased array based on four-dimensional joint weighting and dynamic pre-switching.

[0009] To achieve the above object, the technical solution adopted by the present invention is:

[0010] A phased array tracking method based on four-dimensional joint weighted dynamic pre-switching, which is applied to a drone phased array system, and includes the following steps:

[0011] Step a, receiving spatio-temporal correlation parameters through a target trajectory prediction module, including the target azimuth angle , elevation angle and speed , update the time-space correlation parameters to the target historical motion trajectory database, generate the target state prediction value based on the UKF prediction method, and send a trigger signal to the array switching control module;

[0012] Step b: after receiving the trigger signal, the array switching control module starts the four-dimensional joint gradual weighting program, performs linear phase interpolation in the delay dimension, generates a phase-continuous control signal, and sends a mutual coupling compensation request to the phase calibration module;

[0013] Step c: after receiving the mutual coupling compensation request, the phase calibration module solves the mutual coupling matrix, establishes a cross-array phase mapping model based on cubic spline interpolation, compensates the residual phase error to <10°, and sends a phase calibration completion signal to the beamforming module;

[0014] Step d, the beamforming module uses the real-time radar scattering cross section Dynamically adjust the number of effective array elements involved in beamforming , the optimization criterion is to maximize the signal-to-noise ratio ,in is the Boltzmann constant, is the system noise temperature, is the instantaneous signal bandwidth, is the system loss factor; according to the number of effective array elements The final beam pointing signal is generated according to the dynamic adjustment rules and sent to the antenna control unit to complete the stable relay tracking of the current cycle.

[0015] Furthermore, the state equation of the UKF prediction method described in step a is:

[0016] ,

[0017] Among them, the state vector , is the state transfer matrix, is zero-mean Gaussian noise, represents the change of the corresponding parameter, and the superscript T represents the transpose of the matrix;

[0018] Process noise covariance matrix of UKF prediction method The update rules are:

[0019] ,

[0020] in, For the forgetting factor; for Prediction value of the state at the moment; represents the discretized time step index, corresponding to the system's duty cycle or sampling moment; Denote the prior state estimate value, that is, based on the moment and all the observation data before it, for the prediction of the state at the moment.

[0021] Furthermore, the specific manner of the four-dimensional joint gradient weighting program described in step b is as follows:

[0022] (1) After the front-end switching control module receives the trigger signal, it obtains the real-time measurement values of the current beam dwell period T and the adjacent front-end polarization axis offset angle θ, and initializes the time counter t = 0; the beam dwell period refers to the duration that the beam stays at a single pointing wave position;

[0023] (2) According to the polarization axis offset angle measured in real time, construct the rotation matrix in the polarization dimension to align the polarization directions of the adjacent front-end antenna elements in phase;

[0024] (3) Within the time interval , synchronously calculate the exit front-end weight function and the entry front-end weight function , where t increases in integer multiples of the pulse repetition period , and is reset to zero when t reaches T;

[0025] (4) Map and to the transmission power ratios of the exit front-end and the entry front-end respectively, with the constraint condition to achieve continuous transition of the radiation energy during the switching process;

[0026] (5) Based on the alignment result in step (2), the two weight functions in step (3), and the transmission power ratio in step (4), calculate the front-end switching angle in the spatial dimension:

[0027] ,

[0028] ,

[0029] Generate a four-dimensional joint weighted phase control signal;

[0030] (6) Send the phase control signal generated in step (5) to the phase calibration module in the form of a mutual coupling compensation request.

[0031] Furthermore, the measurement method of the polarization axis offset angle is as follows:

[0032] (1) Generate a polarization phase difference histogram through the polarization direction calibration signal of the array antenna element;

[0033] (2) The least squares algorithm is used to fit the peak points of the histogram, and the sum of the squared residual phase differences is minimized through iterative optimization to calculate the angle between the actual polarization axis and the theoretical axis. .

[0034] Further, in step c, the mutual coupling matrix is solved in the following specific manner:

[0035] (1) Receive the signals of the geosynchronous beacon satellites with azimuth , elevation .

[0036] (2) Construct an overdetermined system of equations based on the cross-correlation function of the signals received by multiple arrays:

[0037] ,

[0038] where x is the mutual coupling parameter vector, obtained through basis function expansion and least squares solution, and is used to compensate for the electromagnetic coupling effect between array elements; b is the observed signal vector, generated from the signals of the geosynchronous beacon satellites received by multiple arrays, and reflects the coupling between array elements and the difference in propagation paths;

[0039] (3) Calculate the least squares solution through the SVD decomposition method to obtain:

[0040] .

[0041] Further, in step c, a cross-array phase mapping model is established based on cubic spline interpolation in the following specific manner:

[0042] In the adjacent array switching region, a cross-array phase mapping model is established using piecewise cubic spline functions, and the phase function in each sub-interval is expressed as:

[0043]

[0044] where is the phase mapping function in the th array switching interval, in radians; is the spatial angle parameter of the array switching, in degrees; is the starting switching angle of the th array, measured in real time by the array installation attitude sensor; are the cubic spline interpolation coefficients.

[0045] Further, the dynamic adjustment rule for the number of effective array elements in step d is as follows:

[0046] (1) When ≥ 10 dBsm, the lower limit of the activation element ratio is 30% of the total number of array elements;

[0047] When the signal-to-noise ratio SNR < 10 dB, automatically switch to the full-array working mode;

[0048] The adjustment step of the element activation ratio is 5% of the total number of elements.

[0049] Furthermore, it also includes the step of phase continuity monitoring:

[0050] (1) Adopt the dual-frequency point comparison method to synchronously transmit two detection signals with different wavelengths ;

[0051] (2) Calculate the equivalent wavelength , and amplify the measurement accuracy of the phase jump;

[0052] When it is detected that the differential phase jump during the switching process exceeds , trigger the four-dimensional joint gradual weighting program, as the working wavelength.

[0053] The beneficial effects of the present invention are as follows:

[0054] 1. High-precision phase continuity: The present invention can reduce the RMS value of the residual phase error during array switching through four-dimensional joint gradual weighting (azimuth / pitch / polarization / time delay) and cubic spline phase mapping model, meeting the high-precision tracking requirements.

[0055] 2. Improved dynamic response speed: The present invention can predict the switching timing N cycles in advance through the UKF prediction algorithm, reducing the tracking interruption time.

[0056] 3. Optimization of energy efficiency ratio: The present invention's adaptive sub-array reconstruction based on RCS can reduce the system power consumption while maintaining a relatively high communication quality. Specific implementation manner

[0057] Next, the technical solution of the present invention will be further described. Obviously, these contents are only a part of the embodiments of the present invention, rather than all embodiments. Based on the following embodiments, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0058] A phased array tracking method based on four-dimensional joint weighted dynamic pre-switching is applied to an unmanned aerial vehicle phased array system. During the transition period between adjacent array switching, four-dimensional joint gradual weighting of azimuth, pitch, polarization, and time delay is synchronously implemented for the exiting array and the entering array, and the transition time is set to beam dwell cycles, where

[0059] Step a: The target trajectory prediction module receives spatio-temporal correlation parameters, including the target azimuth angle , elevation angle , and velocity . Update the spatio-temporal correlation parameters to the target historical motion trajectory database, generate a target state prediction value based on the UKF prediction method, and send a trigger signal to the array switching control module. Among them, the state equation of the UKF prediction method is:

[0060] ,

[0061] where the state vector , is the state transition matrix, is zero-mean Gaussian noise, represents the change in the corresponding parameter, and the superscript T represents the transpose of the matrix;

[0062] The process noise covariance matrix of the UKF prediction method is updated as follows:

[0063] ,

[0064] where is the forgetting factor, which can be taken as 0.95 in this example; is the state prediction value at time; represents the discretized time step index, corresponding to the working cycle or sampling time of the system; represents the prior state estimate value, that is, based on time and all the observation data before that, the prediction of the state at time.

[0065] Step b: After receiving the trigger signal, the array switching control module starts the four-dimensional joint gradual weighting program, performs linear phase interpolation in the time delay dimension, generates a phase-continuous control signal, and sends a mutual coupling compensation request to the phase calibration module. The specific method of the four-dimensional joint gradual weighting program is as follows:

[0066] (1) After receiving the trigger signal, the array switching control module obtains the current beam dwell period T and the real-time measurement value of the adjacent array polarization axis offset angle , and initializes the time counter t = 0. The beam dwell period refers to the duration that the beam stays at a single pointing wave position. The measurement method of the polarization axis offset angle is as follows:

[0067] (1.1) Generate a polarization phase difference histogram through the polarization direction calibration signal of the array antenna unit;

[0068] (1.2) The peak points of the histogram are fitted using the least - squares algorithm, and the sum of the squared residual phase differences is minimized through iterative optimization to calculate the angle between the actual polarization axis and the theoretical axis. .

[0069] (2) Based on the polarization axis offset angle measured in real - time , a rotation matrix in the polarization dimension is constructed to align the polarization directions of the antenna elements on adjacent array surfaces in terms of phase;

[0070] In this example, the test frequency band is the C - band, the target speed is 200 m / s, the array configuration is a four - face phased array, with 1024 elements per array surface, and the polarization axis offset angle (measured value). The test results are shown in the following table:

[0071]

[0072] (3) Within the time interval , the weight function of the exiting array surface and the weight function of the entering array surface are calculated synchronously, where t increases in integer multiples of the pulse repetition period and is reset to zero when t reaches T;

[0073] (4) Map and to the transmission power ratios of the exiting array surface and the entering array surface respectively. The constraint condition is to achieve a continuous transition of the radiated energy during the handover process; specifically:

[0074] Transmission power ratio of the exiting array surface: ,

[0075] Transmission power ratio of the entering array surface: ,

[0076] where , ranges from beam dwell periods.

[0077] (5) Based on the alignment result in step (2), the two weight functions in step (3), and the transmission power ratios in step (4), calculate the array surface handover angle in the spatial dimension:

[0078] ,

[0079] ,

[0080] Generate a four - dimensional jointly weighted phase control signal;

[0081] During the phase control process, the rotation matrix form in the polarization dimension is as follows:

[0082] ,

[0083] The adjustment step accuracy of is 0.1°.

[0084] (6) Send the phase control signal generated in step (5) to the phase calibration module in the form of a mutual coupling compensation request.

[0085] Step c: After receiving the mutual coupling compensation request, the phase calibration module solves the mutual coupling matrix, establishes a cross-array phase mapping model based on cubic spline interpolation, compensates the residual phase error to <10°, where the residual phase error includes the mutual coupling phase distortion introduced by array switching, and sends a phase calibration completion signal to the beam synthesis module; the specific method for solving the mutual coupling matrix is as follows:

[0086] (1) Receive the synchronous orbit beacon satellite signals of the azimuth angle , elevation angle ;

[0087] (2) Construct an overdetermined system of equations based on the cross-correlation function of the multi-array received signals:

[0088] ,

[0089] where x is the mutual coupling parameter vector, obtained by basis function expansion and least squares solution, and is used to compensate the electromagnetic coupling effect between array elements; b is the observation signal vector, generated by receiving the synchronous orbit beacon signals by the multi-array, and reflects the coupling between array elements and the propagation path difference;

[0090] (3) Calculate the least squares solution through the SVD decomposition method to obtain:

[0091] .

[0092] Establish a cross-array phase mapping model based on cubic spline interpolation, and the specific method is as follows:

[0093] In the adjacent array switching area, use a piecewise cubic spline function to establish a cross-array phase mapping model, and the phase function in each sub-interval is expressed as:

[0094]

[0095] where is the phase mapping function in the th array switching interval, with the unit of radian; is the spatial angle parameter of array switching, with the unit of degree; is the The starting switching angle of each array face is measured in real time by an array face installation attitude sensor; is the cubic spline interpolation coefficient, and its value satisfies the following conditions:

[0096] 1) Phase continuity: , where is the measured phase reference value obtained through polarization calibration and mutual coupling compensation at the starting angle of array face switching;

[0097] 2) First derivative continuity: ;

[0098] 3) Second derivative continuity: .

[0099] Step d, the beam synthesis module dynamically adjusts the number of effective array elements participating in beam synthesis according to the real-time radar cross section , and the optimization criterion is to maximize the signal-to-noise ratio , where is the Boltzmann constant, is the system noise temperature, is the instantaneous signal bandwidth, is the system loss factor; according to the dynamic adjustment rule of the number of effective array elements , generate the final beam pointing signal and send it to the antenna control unit to complete the stable handover tracking of the current cycle; the dynamic adjustment rule of the number of effective array elements is as follows:

[0100] (1) When ≥10 dBsm, the lower limit of the activation ratio of array elements is 30% of the total number of array elements;

[0101] (2) When the signal-to-noise ratio SNR < 10 dB, automatically switch to the full array face working mode;

[0102] The adjustment step size of the array element activation ratio is 5% of the total number of array elements.

[0103] In one embodiment, a 5% adjustment step size comparison experiment is carried out. The scenario is =12 dBsm, =10 ms, the target distance is 50 km, and the variable is the adjustment step size (2% / 5% / 10%). Evaluate the beam pointing error and reconstruction time, and the experimental results are shown in Table 2 below:

[0104]

[0105] It can be seen that when the step size > 5%, the beam pointing error increases by 1.2 dB, and when the step size < 5%, the reconstruction time exceeds by 30%.

[0106] Step e, phase continuity monitoring, specifically:

[0107] (1) Adopt a dual-frequency point comparison method to synchronously transmit two detection signals with different wavelengths ;

[0108] (2) Calculate the equivalent wavelength to amplify the measurement accuracy of phase jumps;

[0109] (3) When it is detected that the differential phase jump during the switching process exceeds , trigger a four-dimensional joint gradual weighting program, being the working wavelength.

[0110] In view of the problems of poor phase continuity, insufficient dynamic response, limited calibration accuracy, and low energy efficiency ratio in conventional phased array systems, the present invention solves the problems of phase jumps, prediction lags, and energy consumption in multi-array switching through methods such as four-dimensional joint gradual weighting, dynamic pre-switching mechanism, cross-array phase calibration technology, and adaptive sub-array reconstruction strategy. It is applicable to the continuous tracking scenario of high-speed moving targets and can be used in vehicle-mounted phased array platforms.

[0111] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A phased array tracking method based on four-dimensional joint weighted dynamic pre-switching, applied to a UAV phased array system, characterized in that: The steps include: Step a, receiving the time-space correlation parameters, including the target azimuth, through the target trajectory prediction module , Pitch angle and speed , update the time-space correlation parameters to the target historical motion trajectory database, generate the target state prediction value based on the UKF prediction method, and send a trigger signal to the array switching control module; Step b: after receiving the trigger signal, the array switching control module starts the four-dimensional joint gradual weighting program, performs linear phase interpolation in the delay dimension, generates a phase-continuous control signal, and sends a mutual coupling compensation request to the phase calibration module; Step c: after receiving the mutual coupling compensation request, the phase calibration module solves the mutual coupling matrix, establishes a cross-array phase mapping model based on cubic spline interpolation, compensates the residual phase error to <10°, and sends a phase calibration completion signal to the beamforming module; Step d, the beamforming module uses the real-time radar scattering cross section Dynamically adjust the number of effective array elements involved in beamforming , the optimization criterion is to maximize the signal-to-noise ratio ,in is the Boltzmann constant, is the system noise temperature, is the instantaneous signal bandwidth, is the system loss factor; according to the number of effective array elements The final beam pointing signal is generated according to the dynamic adjustment rules and sent to the antenna control unit to complete the stable relay tracking of the current cycle.

2. The phased array tracking method based on four-dimensional joint weighted dynamic pre-switching according to claim 1 is characterized in that: The equation of state for the UKF prediction method described in step a is: ， Among them, the state vector , is the state transfer matrix, is zero-mean Gaussian noise, represents the change of the corresponding parameter, and the superscript T represents the transpose of the matrix; Process noise covariance matrix of UKF prediction method The update rules are: ， in, For the forgetting factor; for Prediction value of the state at the moment; represents the discretized time step index, corresponding to the system's duty cycle or sampling moment; Represents the prior state estimate, that is, based on All observation data before and after time Prediction of the state at a given moment.

3. The phased array tracking method based on four-dimensional joint weighted dynamic pre-switching according to claim 1, characterized in that: The specific method of the four-dimensional joint gradient weighting procedure described in step b is: (1) After receiving the trigger signal, the array switching control module obtains the real-time measurement value of the current beam dwell period T and the adjacent array polarization axis offset angle θ, and initializes the time counter t=0; the beam dwell period refers to the duration of the beam staying at a single pointing wave position; (2) Based on the real-time measurement of the polarization axis offset angle , construct the rotation matrix of the polarization dimension , phase align the polarization directions of adjacent array antenna units; (3) In the time interval Inside, synchronously calculate the exit front weight function and the cutting front weight function , where t is the pulse repetition period Increment by integer multiples and reset to zero when t reaches T; (4) and The transmission power ratios mapped to the exit and entry planes are respectively, and the constraints are , to achieve continuous transition of radiation energy during switching; (5) Based on the alignment result of step (2), the two weight functions of step (3) and the transmit power ratio of step (4), the array switching angle is calculated in the spatial dimension: ， ， generating a four-dimensional jointly weighted phase control signal; (6) The phase control signal generated in step (5) is sent to the phase calibration module in the form of a mutual coupling compensation request.

4. The phased array tracking method based on four-dimensional joint weighted dynamic pre-switching according to claim 3 is characterized in that: The polarization axis offset angle The measurement method is: (1) Generate a polarization phase difference histogram by calibrating the polarization direction signal of the array antenna unit; (2) The least squares algorithm is used to fit the histogram peak point, and the sum of squares of the residual phase difference is minimized through iterative optimization to solve the angle between the actual polarization axis and the theoretical axis. .

5. The phased array tracking method based on four-dimensional joint weighted dynamic pre-switching according to claim 1, characterized in that: In step c, the mutual coupling matrix is ​​solved in the following way: (1) Receiving azimuth , Pitch angle The synchronous orbit beacon satellite signal; (2) Construct an overdetermined set of equations based on the cross-correlation function of the multi-array received signals: ， Among them, x is the mutual coupling parameter vector, which is obtained by basis function expansion and least square solution, and is used to compensate for the electromagnetic coupling effect between array elements; b is the observation signal vector, which is generated by multiple arrays receiving synchronous orbit beacon signals, reflecting the coupling between array elements and the difference in propagation paths; (3) Calculate the least squares solution using the SVD decomposition method and obtain: 。 6. The phased array tracking method based on four-dimensional joint weighted dynamic pre-switching according to claim 5, characterized in that: In step c, a cross-plane phase mapping model is established based on cubic spline interpolation, specifically in the following way: In the adjacent array switching area, the piecewise cubic spline function is used to establish the cross-array phase mapping model. The phase function in each sub-interval is expressed as: in, For the The phase mapping function within the switching interval of each array, in radians; It is the spatial angle parameter of the array switching, in degrees; For the The starting switching angle of each array is measured in real time by the attitude sensor installed on the array; are cubic spline interpolation coefficients.

7. The phased array tracking method based on four-dimensional joint weighted dynamic pre-switching according to claim 1, characterized in that: The number of effective array elements in step d The dynamic adjustment rules are: (1) When When ≥10 dBsm, the lower limit of the ratio of activated array elements is 30% of the total number of array elements; (2) When the signal-to-noise ratio (SNR) is less than 10dB, it automatically switches to the full array working mode; The adjustment step of the array element activation ratio is 5% of the total number of array elements.

8. The phased array tracking method based on four-dimensional joint weighted dynamic pre-switching according to claim 1, characterized in that: Also included are the steps for phase continuity monitoring: (1) Using dual-frequency comparison method, synchronously transmitting two different wavelengths Detection signal; (2) Calculation of equivalent wavelength , amplify the measurement accuracy of phase jump; (3) When the differential phase jump exceeds When , the four-dimensional joint gradient weighting procedure is triggered, is the working wavelength.

Images