CN-121980806-A - Polarization sensitive array parameter low-complexity real-time estimation method by using external reference source

CN121980806ACN 121980806 ACN121980806 ACN 121980806ACN-121980806-A

Abstract

The invention belongs to the technical field of polarization sensitive array parameter estimation, and particularly relates to a low-complexity real-time estimation method for polarization sensitive array parameters by utilizing an external reference source, wherein a single full-component electromagnetic vector sensor is dummied into an array through a time interval, the sensor receives three-component signals of an electric field and a magnetic field, a narrow-band full-polarization signal source is selected, carrier frequency, sampling frequency, snapshot number and signal source real parameters are set, and Monte Carlo simulation times and signal to noise ratio are set; and receiving an overlapped signal containing channel amplitude and phase errors and an priori standard source signal through simulation modeling, wherein parameters of the overlapped signal are unknown, only signal data are known, parameters of the priori standard source signal and a signal model are known, and calculating a guide matrix by utilizing parameters of the priori standard source signal to obtain the priori standard source signal without errors and noise.

Inventors

WU MENGJIE
BI XIAOWEN
GAO YUAN
SI CHENGKE
HU YAN
FAN RONG

Assignees

中国民用航空飞行学院

Dates

Publication Date: 20260505
Application Date: 20260202

Claims (10)

1. The low-complexity real-time estimation method for the polarization sensitive array parameters by using the external reference source is characterized by comprising the following steps of: The method comprises the steps that a single full-component electromagnetic vector sensor is dummied into an array through a time interval, the sensor receives three-component signals of an electric field and a magnetic field, a narrow-band full-polarization signal source is selected, carrier frequency, sampling frequency, snapshot number and real parameters of the signal source are set, and Monte Carlo simulation times and signal to noise ratio are set; step two, receiving an overlapped signal containing channel amplitude and phase errors and a priori standard source signal through simulation modeling, wherein parameters of the overlapped signal are unknown and only signal data are known, and parameters and a signal model of the priori standard source signal are known; Calculating a guide matrix by using the prior standard source signal parameters to obtain an error-free and noise-free prior standard source signal, dividing the two types of prior standard source signals into M sections, estimating the error of each section of channel by least square, averaging the sections to obtain an error estimation matrix, and substituting the error estimation matrix into a target signal to calculate a target receiving signal after error correction; step four: Before the extraction of the Chinese medicine Individual snapshots as Extracting time delay points After that Individual snapshots as . Will be And Input matrix for acquiring ESPRIT algorithm by splicing ; Fifthly, carrying out signal subspace tracking on an input matrix Z by adopting an online projection approximate subspace tracking algorithm, and obtaining the signal subspaces of two divided adjacent subarrays through algorithm parameter initialization, projection matrix iterative updating and orthogonalization processing And The two satisfy the rotation invariable relation; Step six, utilizing And Solving a rotation matrix according to the rotation invariant relation of the guide matrix, decomposing the eigenvalue of the rotation matrix, and calculating to obtain an estimated value of the guide matrix; Step seven, constructing a Potentilla vector through an electromagnetic field component based on the guide matrix estimated value, calculating an azimuth angle and a pitch angle, and reversely solving a polarization auxiliary angle and a polarization phase difference by using a polarization response model; And step eight, repeating the step two to the step seven for multiple Monte Carlo simulation, calculating the root mean square error of each parameter, and evaluating the stability and the estimation precision of the algorithm.
2. The method for low-complexity real-time estimation of polarization-sensitive array parameters using external reference sources of claim 1, wherein the dummy array in step one is implemented by: Taking the first 1024 snapshots as the signal data of the first virtual array element, intercepting the signal data of the second virtual array element with the same length after 200 snapshots are separated, and converting the time interval into a physical interval.
3. The method for estimating the polarization sensitive array parameters with low complexity in real time by using the external reference source according to claim 1, wherein the specific process of estimating the channel errors in the third step is as follows: ; Wherein, the In order to obtain the number of segments, Carrying out least square estimation on each section of prior standard source signal and the prior standard source signal without error noise to obtain a segmented channel error, and obtaining an error estimation matrix through section average calculation, wherein the formula is as follows: ; Wherein the method comprises the steps of Is the channel error estimate for the mth segment.
4. The method for low-complexity real-time estimation of polarization-sensitive array parameters using external reference sources of claim 1, wherein the initializing parameters of the online projection approximation subspace tracking algorithm in step five comprises: setting forgetting factor as 0.99, generating random Gaussian matrix, orthogonalizing to obtain initial subspace orthogonalization basis Initializing a projection matrix as a dimensionality-matched identity matrix.
5. The method for estimating the polarization-sensitive array parameters in real time with low complexity by using the external reference source according to claim 1, wherein in the iterative updating process of the projection matrix in the fifth step, the iteration confirmation period is set to be 5, the orthogonalization process is performed once for each 5 iterations on the current subspace orthogonalization base, and the QR orthogonalization is performed on the final subspace orthogonalization base after all iterations are finished.
6. The method for low-complexity real-time estimation of polarization-sensitive array parameters using external reference sources as set forth in claim 1, wherein the signal subspaces of two adjacent subarrays in step five And The dividing mode of (a) is as follows: The first 6 rows of the final subspace orthogonal basis are taken as The last 6 rows are taken as Both satisfy Wherein Is a rotation matrix.
7. The method for estimating the polarization-sensitive array parameters in real time with low complexity by using the external reference source as set forth in claim 1, wherein the step six is characterized in that the calculation process of the guide matrix estimated value is as follows: By passing through Solving a rotation matrix For a pair of Decomposing the eigenvalue to obtain eigenvalue matrix, and deriving to obtain guide matrix estimated value 。
8. The method for estimating the polarization-sensitive array parameters in real time with low complexity by using the external reference source according to claim 1, wherein the calculating process of the azimuth angle and the pitch angle in the seventh step is as follows: Extracting electric field vector of signal source from guide matrix estimated value And magnetic field vector After normalization processing, a propagation direction unit vector is constructed Then pass through Calculating pitch angle By means of Calculating azimuth angle 。
9. The method for estimating the polarization sensitive array parameters in real time with low complexity by using the external reference source according to claim 1, wherein the inverse solution of the polarization auxiliary angle and the polarization phase difference in the seventh step is as follows: constructing a polarization transformation matrix T based on the estimated azimuth angle and pitch angle, and solving a polarization coefficient vector by a least-squares method ; ; Then pass through Polarization auxiliary angle By means of Polarization phase difference 。
10. The method for estimating the polarization sensitive array parameter with low complexity in real time by using the external reference source as claimed in claim 1, wherein in the first step, the Monte Carlo simulation times are set to 200 times, and the signal to noise ratio is set to 15dB; in the seventh step, the root mean square error of each parameter is controlled to be 1 degree.

Description

Polarization sensitive array parameter low-complexity real-time estimation method by using external reference source Technical Field The invention belongs to the technical field of polarization sensitive array parameter estimation, and particularly relates to a low-complexity real-time estimation method for polarization sensitive array parameters by using an external reference source. Background In the field of polarization sensitive array parameter estimation, accurate acquisition of azimuth angle, pitch angle and polarization parameters is a core requirement of scenes such as radar target detection, communication signal reception, aviation response signal analysis and the like. However, in practical application, the array is susceptible to channel amplitude-phase errors, multi-signal source overlapping interference and environmental noise disturbance, so that the traditional parameter estimation method faces significant challenges, such as DOA estimation method or covariance matrix inversion operation based on feature decomposition, high dimension is required to be processed, calculation complexity is high, real-time processing scene is difficult to adapt, and meanwhile, when non-homologous signal overlapping conditions are faced, the precision and stability of parameter estimation are easy to be greatly reduced, and the dual requirements of high efficiency and accuracy in engineering application cannot be met. In order to reduce complexity, the prior art has developed multi-direction exploration, wherein part of methods simplify parameter space through sparse representation, but redundancy easily occurs in dictionary construction when multi-parameter joint estimation of a polarization sensitive array is faced, so that memory occupation is overlarge and iteration efficiency is low; in addition, the method aims at the non-uniform noise environment optimization angle estimation, but does not fully consider the problem of cooperative estimation of polarization parameters and space parameters, and lacks a targeted correction mechanism for channel amplitude-phase errors, and even in SAR image scattering center parameter estimation, complexity optimization is realized through image domain segmentation and frequency domain fusion, but the technical idea is difficult to directly migrate to a real-time signal processing scene of a polarization sensitive array, so that a polarization sensitive array parameter estimation scheme which can integrate an external reference source to realize error correction and simultaneously has low complexity and high real-time performance is needed. Disclosure of Invention The invention aims at solving the technical problems and provides a low-complexity real-time estimation method for polarization sensitive array parameters by using an external reference source. In view of the above, the present invention provides a low-complexity real-time estimation method for polarization sensitive array parameters by using external reference sources, comprising the following steps: The method comprises the steps that a single full-component electromagnetic vector sensor is dummied into an array through a time interval, the sensor receives three-component signals of an electric field and a magnetic field, a narrow-band full-polarization signal source is selected, carrier frequency, sampling frequency, snapshot number and real parameters of the signal source are set, and Monte Carlo simulation times and signal to noise ratio are set; step two, receiving an overlapped signal containing channel amplitude and phase errors and a priori standard source signal through simulation modeling, wherein parameters of the overlapped signal are unknown and only signal data are known, and parameters and a signal model of the priori standard source signal are known; Calculating a guide matrix by using the prior standard source signal parameters to obtain an error-free and noise-free prior standard source signal, dividing the two types of prior standard source signals into M sections, estimating the error of each section of channel by least square, averaging the sections to obtain an error estimation matrix, and substituting the error estimation matrix into a target signal to calculate a target receiving signal after error correction; step four: Before the extraction of the Chinese medicine Individual snapshots asExtracting time delay pointsAfter thatIndividual snapshots as. Will beAndInput matrix for acquiring ESPRIT algorithm by splicing; Fifthly, adopting an online projection approximate subspace tracking algorithm to input a matrixSignal subspace tracking is carried out, and the signal subspaces of the two divided adjacent subarrays are obtained through algorithm parameter initialization, projection matrix iteration updating and orthogonalization processingAndThe two satisfy the rotation invariable relation; Step six, utilizing AndSolving a rotation matrix according to the rotation invariant relation of the gu