Polarization generalized likelihood ratio test method for joint search in spatial and polarization domains

Abstract

The present invention relates to the field of antenna signal processing technology, and more specifically, to a spatial-polarization domain joint search polarization generalized likelihood ratio test method applicable to arbitrary polarization-sensitive arrays and coherent and incoherent signal environments. Compared with a traditional minimum description length method, the method has higher signal source number estimation performance under low signal-to-noise ratio and low snapshot number conditions. The method models the signal source number estimation as a test problem, using 0 and 1 to represent the presence or absence of a signal. For a signal source greater than M, M signal source tests are required to complete the estimation. The method specifically comprises the following steps: calculating a likelihood ratio L1(x), then comparing the likelihood ratio L1(x) with a predetermined threshold μ1; if the value is not greater than the threshold, stopping the search, and updating the signal source number estimate #imgabs0#; otherwise, updating #imgabs1# using a maximum likelihood estimation method.

Description

Technical field:

[0001] The present invention relates to the field of antenna signal processing technology, and more specifically to a spatial-polarization domain joint search polarization generalized likelihood ratio test method applicable to arbitrary polarization-sensitive arrays and coherent and incoherent signal environments. Compared with the traditional minimum description length method, the method has higher signal source number estimation performance under low signal-to-noise ratio and low snapshot number conditions. Background technology:

[0002] Conformal array antennas can be attached to carrier surfaces to improve space utilization. Their three-dimensional curved structure increases spatial freedom, thereby achieving greater spatial beam coverage, larger array effective aperture, and antenna gain. Estimating DOA-polarization parameters based on conformal arrays is fundamental research for furthering conformal array engineering applications. However, most current DOA estimation methods assume a known number of signal sources, a crucial issue in practical applications.

[0003] Currently, most coherent signal source number estimation methods require arrays of specific shapes, such as uniform linear arrays and uniform circular arrays. However, in practice, due to errors such as mutual coupling of array elements, array element position errors, and near-field reflections from the installation platform, the actual array exhibits irregular, non-ideal, and non-periodic spatial distribution characteristics. Therefore, it is necessary to study source number estimation methods suitable for arbitrary arrays.

[0004] Currently, there is limited research on estimating the number of coherent sources in arbitrary arrays. The minimum description length method is not restricted by array configuration and can estimate the number of coherent signal sources, but the search complexity is too high under multiple incident signals. The generalized likelihood ratio test method, based on the principle of sequential hypothesis testing, is suitable for arbitrary scalar arrays and coherent and incoherent signal environments. It has good performance under low signal-to-noise ratios, but cannot detect polarization signals. In practice, the number of sources in conformal arrays is a prerequisite for most DOA-polarization parameter estimation algorithms. Therefore, studying the problem of source number estimation under the condition of conformal arrays without array constraints is of great importance. Summary of the invention:

[0005] In order to solve the problem that traditional signal source number estimation methods are limited in array configuration and fail in coherent signal source environments, the present invention extends the generalized likelihood ratio test method applicable to scalar arrays to vector arrays of polarization-sensitive arrays, and proposes a polarization generalized likelihood ratio test method that is applicable to any polarization-sensitive array and can estimate coherent signals and incoherent signals through joint search in the spatial and polarization domains.

[0006] The present invention is achieved by the following measures:

[0007] A method for joint search of polarization generalized likelihood ratio in spatial and polarization domains is characterized in that the number of signal sources is modeled as a test problem, with 0 and 1 representing the presence or absence of a signal. For the case of M>1 signal sources, M signal source tests are required to complete the estimation. For the mth search, the DOA-polarization parameter needs to be known. The first m-1 incident signals can determine the DOA-polarization parameter: The existence or non-existence of the mth source of information, the hypothesis testing problem of the mth stage is modeled as:

[0008]

[0009] in Assume that there is a signal in the first m-1 stages, but no signal in the mth stage; Assume that there is a signal in the first m-1 stages and there is also a signal in the mth stage; the spatial domain and polarization domain joint search polarization generalized likelihood ratio test method includes the following steps:

[0010] Step 1: Calculate the likelihood ratio L1(x), and then compare it with the predetermined threshold μ1. If the value is not greater than the threshold, stop searching and estimate the number of sources. Otherwise, use the maximum likelihood estimation method to update

[0011] Step 2: When m>1, A m-1 for where a m-1 For the The steering vector of the signal;

[0012] Step 3: According to A m-1 Compute projection operator And calculate the likelihood ratio L m (x);

[0013] Step 4: Calculate the L m (x) compared with the threshold value of the mth stage;

[0014] Step 5: If L m (x) is less than the threshold, the search is terminated, and the estimated number of sources is Estimated parameters

[0015] Step 6: If L m (x) is greater than the threshold, and all estimated incident angle parameters are updated using the maximum likelihood estimation method.

[0016] The likelihood ratio L1(x) described in step 1 of the present invention is derived as follows: Assuming that the noise variance is known, the likelihood ratio applicable to the scalar array is generalized to the polarized signal to obtain the likelihood ratio:

[0017]

[0018] Where μ m is the mth level threshold determined based on the mth level false alarm probability; a m is the estimated parameter The corresponding N×1-dimensional steering vector; K is the number of snapshots; is the projection operator:

[0019]

[0020] So for m=1, the likelihood ratio estimate is:

[0021]

[0022] The threshold determination in step 1 of the present invention is very important, as it determines the success rate of the source number estimation. Simplifying the likelihood ratio can yield

[0023]

[0024] Define the random variable z(t)

[0025]

[0026] Where z(t); t=1,...,K are K independent normally distributed complex random variables with mean μ m (t) The unit variable is, where μ m (t) is given by:

[0027]

[0028] A described in step 2 of the present invention m-1 It is the spatial-polarization array flow pattern of m-1 signals in M×(m-1) dimensions:

[0029]

[0030] in is a (m-1)×1 dimensional signal, s m (t) is the time domain waveform of the mth incident signal, and t is the sampling time.

[0031] The spatial-polarization domain joint search polarization generalized likelihood ratio test method proposed in the present invention is applicable to any polarization-sensitive array and to coherent and incoherent signal environments. Compared with the traditional minimum description length method, it has higher source number estimation performance under low signal-to-noise ratio and low snapshot number conditions.

[0032] The model used in the present invention is a polarization conformal array signal model, and the polarization-spatial representation of the electromagnetic wave of the model is as follows:

[0033] Consider a fully polarized electromagnetic wave incident signal incident on an array element. The incident direction of the electromagnetic wave signal can be determined only by the azimuth and elevation angles. Let θ be the azimuth angle, and θ is defined as the angle between the projection of the incident signal on the xoy plane and the positive direction of the x-axis. is the pitch angle, Defined as the angle between the incident signal and the positive direction of the z-axis; the polarization of the electromagnetic wave can describe the trajectory of the electric field endpoint propagation in space, according to the two electric field components perpendicular to the propagation direction of the incident signal and The polarization state is determined by the amplitude ratio and phase difference of the polarization state, and γ is defined as the polarization auxiliary angle. and The amplitude ratio of the components; η is the polarization phase difference, η is defined as and The phase difference of the components is assumed to be a fully polarized wave incident from the narrowband far field. The incident electromagnetic wave x(t) can be expressed as:

[0034]

[0035] Where f0 is the carrier frequency; k = -(2π / λ) is the propagation wave number; is the propagation direction; s(t) is the complex baseband signal; E h =cosγ is the horizontal electric field strength, E v =sinγe jη is the electric field intensity in the vertical direction; the electric field intensity component of the electromagnetic wave is obtained by using the conversion relationship between the rectangular coordinate system and the spherical coordinate system, [E r ,E h ,E v ] T is the electric field intensity component in spherical coordinates, [E x ,E y ,E z ] T is the electric field intensity component in rectangular coordinates, E r If is 0, then:

[0036]

[0037] Similarly, the magnetic field intensity component can be obtained, and the electromagnetic field vector of the fully polarized wave in the rectangular coordinate system is:

[0038]

[0039] The model of the present invention can also be a polarization conformal array signal model, and the traditional steering vector model of the model is as follows:

[0040] Because the curvature of the carrier causes conformal array elements to change their pattern gain during installation, traditional signal modeling based on consistent element pattern gain is no longer applicable. For the same incident signal, the azimuth and elevation angles in the local coordinate system of each element vary, resulting in different pattern gains for each element. Therefore, Euler rotation transformations must be used to solve for the steering vector when modeling the received signal of a conformal array.

[0041] Consider an arbitrary conformal array with N elements, and the incident signal is a fully polarized wave in an arbitrary narrowband far field. The received signal data model is:

[0042]

[0043] Where x(t) is the N×1-dimensional received data vector at time t, where N is the number of array elements; s(t) is the M×1-dimensional incident signal vector, where M is the number of signal sources; and N(t) is the N×1-dimensional Gaussian white noise vector. is the spatial-polarization domain steering vector;

[0044]

[0045] In the formula is the spatial steering vector of the conformal array, is the polarization domain steering vector of the conformal array; ω=2π / λ, λ is the wavelength; r n is the position coordinate vector of the nth array element, and the specific expression is r n =[x n ,y n ,z n ];v m is the direction coordinate vector of the mth incident signal in the rectangular coordinate system, which is specifically expressed as Polarization response p mn =h m ·g n is the polarization vector h of the mth signal m In the nth array element pattern g n The projection on the global polar coordinate system is the signal polarization vector h m for:

[0046] h m =h mhE h +h mv E v ,

[0047]

[0048] tanγ=A h / A v ,η=φ h -φ v ,(γ∈[0,π / 2],η∈[0,2π)),

[0049] Where, E h is the horizontal basis vector of the electric field, E v is the basis vector in the vertical direction of the electric field, A h is the horizontal electric field amplitude, A v is the vertical electric field amplitude, φ h is the horizontal electric field phase, φ v is the vertical electric field phase.

[0050] The search time of the spatial-polarization domain joint search polarization generalized likelihood ratio test method of the present invention is less than the search time of the minimum description length method under the condition of a small number of information sources. This is because the spatial-polarization domain joint search polarization generalized likelihood ratio test method needs to judge the threshold each time to determine whether to perform the next search. If the value is not greater than the threshold during the next search, the method is stopped. Under the condition of a small number of information sources, the spatial-polarization domain joint search polarization generalized likelihood ratio test method only needs to search the number of information sources several times to determine the result, while the minimum description length method needs to search all possible results before making a judgment. Therefore, the present invention has a more obvious advantage over the minimum description length method when searching for a low number of information sources. Description of the drawings:

[0051] Attachment Figure 1 Flowchart of the present invention.

[0052] Attachment Figure 2 Schematic diagram of the electromagnetic wave propagation model of the polarization conformal array signal model of the present invention.

[0053] Attachment Figure 3 Schematic diagram of the three-dimensional arbitrary shape array coordinate system of the present invention.

[0054] Attachment Figure 4 (a), (b), and (c) show the impact of the signal-to-noise ratio on the detection success rate of the spatial-polarization domain joint search polarization generalized likelihood ratio test method under the conditions of correlation coefficients of 1, 0.5, and 0, respectively.

[0055] Attachment Figure 5(a), (b), and (c) show the impact of the number of snapshots on the detection success rate of the spatial-polarization domain joint search polarization generalized likelihood ratio test method under the conditions of correlation coefficients of 1, 0.5, and 0, respectively.

[0056] Attachment Figure 6 This is the effect of the number of array elements on the running time of the joint spatial-polarization domain search polarization generalized likelihood ratio test method.

[0057] Attachment Figure 7 (a) and (b) show the effects of the signal-to-noise ratio after position error correction and mutual coupling error correction on the success probability of source number estimation in the spatial-polarization domain joint search polarization generalized likelihood ratio test method. Specific implementation methods:

[0058] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0059] As attached Figure 1 As shown, the present invention proposes a method for joint spatial-polarization domain search and polarization generalized likelihood ratio test. To address the limitations of traditional signal source number estimation methods in array configurations and their failure in coherent signal source environments, the generalized likelihood ratio test method applicable to scalar arrays is extended to vector arrays of polarization-sensitive arrays. This method is applicable to any polarization-sensitive array and can estimate both coherent and incoherent signals. This example specifically includes the following steps:

[0060] The first step is to calculate the likelihood ratio L1(x) and then compare it with the predetermined threshold μ1. If the value is not greater than the threshold, the search is stopped and the number of source estimates is Otherwise, use the maximum likelihood estimation method to update

[0061] The derivation of the likelihood ratio L1(x) is as follows: Assuming the noise variance is known, the likelihood ratio applicable to the scalar array is generalized to the polarized signal to obtain the likelihood ratio:

[0062]

[0063] Where μ m is the mth level threshold determined based on the mth level false alarm probability; a m is the estimated parameter The corresponding N×1-dimensional steering vector; K is the number of snapshots; is the projection operator:

[0064]

[0065] So for m = 1, the likelihood ratio estimate is

[0066]

[0067] The threshold determination is very important, as it determines the success rate of the source number estimation. Simplifying the likelihood ratio yields

[0068]

[0069] Define the random variable z(t)

[0070]

[0071] Where z(t); t=1,…,K are K independent normally distributed complex random variables with mean μ m (t) The unit variable is, where μ m (t) is given by:

[0072]

[0073] Step 2: When m>1, A m-1 for where a m-1 For the The steering vector of the signal. m-1 It is the spatial-polarization array flow pattern of m-1 signals in M×(m-1) dimensions:

[0074]

[0075] in is a (m-1)×1 dimensional signal, s m (t) is the time domain waveform of the mth incident signal, and t is the sampling time.

[0076] The third step is to m-1 Compute projection operator And calculate the likelihood ratio L m (x).

[0077] Assuming the noise variance is known, the likelihood ratio applicable to scalar arrays is extended to polarized signals to obtain the likelihood ratio:

[0078]

[0079] Where μ m is the mth level threshold determined based on the mth level false alarm probability; a m is the estimated parameter The corresponding N×1-dimensional steering vector; K is the number of snapshots; is the projection operator:

[0080]

[0081] So for m = 1, the likelihood ratio estimate is

[0082]

[0083] The fourth step is to calculate L m (x) is compared with the threshold of the mth stage.

[0084] Step 5: If L m If (x) is less than the threshold, the search is terminated. The estimated number of information sources is Estimated parameters

[0085] Step 6: If L m (x) is greater than the threshold, and all estimated incident angle parameters are updated using the maximum likelihood estimation method. Example:

[0086] The performance of the present invention can be illustrated by the following simulation:

[0087] 1. Simulation conditions:

[0088] Simulation 1: Fix the number of incident signal snapshots and observe the impact of different incident signal correlation coefficients on the success rate of source number estimation using the joint spatial-polarization domain search and polarization-based generalized likelihood ratio test method under varying signal-to-noise ratio conditions.

[0089] Simulation conditions: two coherent incident signals with a correlation coefficient of 1. The incident signal parameters are set as follows: The signal frequency is 6GHz, and the number of snapshots is 100. The array parameters are set as follows: a cylindrical conformal array is used, the number of rings is 1, the number of array elements is 20, and they are evenly distributed on a ring with a radius of 0.3m, taking into account the shielding effect. The correlation coefficients of the incident signals are 1, 0.5, and 0, respectively, to observe the effect of the signal-to-noise ratio on the success rate of the signal source number. The false alarm probability of each stage in the simulation in this section is set to 0.01. Because the distribution form of the detection statistic under noise conditions is difficult to express analytically, the detection threshold is obtained through Monte Carlo simulation under given false alarm probability conditions. The detection probability is obtained through 200 Monte Carlo experiments. The simulation results are as follows Figure 4 As shown in (a), (b), and (c).

[0090] Simulation 2: Fixed the incident signal signal-to-noise ratio and observed the impact of different incident signal correlation coefficients on the success rate of source number estimation using the joint spatial-polarization domain search polarization generalized likelihood ratio test method under the condition of varying snapshot number.

[0091] Simulation conditions: two coherent incident signals with a correlation coefficient of 1. The incident signal parameters are set as follows: The signal frequency is 6GHz, and the SNR is -5dB. The array parameters are set as follows: a cylindrical conformal array is used, the number of rings is 1, the number of array elements is 20, and they are evenly distributed on a ring with a radius of 0.3m, taking into account the occlusion effect. The correlation coefficient of the incident signal is 1, 0.5, and 0 respectively, and the effect of the change in signal-to-noise ratio on the success rate of the signal source is observed. The simulation results are shown in Figure 2. Figure 5 As shown in (a), (b), and (c).

[0092] Simulation 3: Record the time required for different methods to estimate the number of signal sources and compare the computational efficiency between algorithms.

[0093] Simulation conditions: Single source Parameter settings: Dual source parameter settings: The signal frequency is 6GHz, the number of snapshots is 200, and the SNR is 10dB. The MATLAB version used in this paper is 2021b, the system is Windows 11, and the CPU model is Intel i7-12700H. The array parameters are set as follows: a cylindrical conformal array is used, the number of rings is 1, and considering the occlusion effect, the array elements are evenly distributed on a ring with a radius of 0.3m. The number of array elements is changed and the single running time of the algorithm is observed. The simulation results of the time required for the algorithm and Figure 6 shown.

[0094] Simulation 4: Analysis of the impact of non-ideal conditions on the success rate of source number estimation.

[0095] Simulation conditions: Observe the performance of the spatial-polarization joint search polarization generalized likelihood ratio test method in the presence of non-ideal factors, that is, the source number estimation performance after position error and mutual coupling error correction. The experimental conditions are the same as those of simulation 1. The simulation results are as follows: Figure 7 As shown in (a) and (b).

[0096] 2. Simulation Results

[0097] from Figure 4 It can be seen from (a), (b), and (c) that the source number detection performance of the spatial-polarization domain joint search polarization generalized likelihood ratio test method is better than that of the minimum description length method at low signal-to-noise ratio (<0dB), and its source number estimation success rate increases with the increase of signal-to-noise ratio.

[0098] from Figure 5 From (a), (b), and (c), we can see that the source number detection performance of the spatial-polarization domain joint search polarization generalized likelihood ratio test method is better than the detection performance of the minimum description length method when the number of snapshots is low (<150), and the success rate of the source number estimation increases with the increase of the number of snapshots. Figure 4 and Figure 5It can be seen that the success rate of source number estimation of the spatial-polarization domain joint search polarization generalized likelihood ratio test method is insensitive to the correlation coefficient of the incident signal, while the minimum description length method is greatly affected by the correlation coefficient.

[0099] Figure 6 The effect of the number of array elements on the running time of the joint search polarization generalized likelihood ratio test method in the spatial and polarization domains is shown in Figure 2. Figure 6 It can be seen from the figure that the running time of the joint search polarization generalized likelihood ratio test method in the spatial-polarization domain under the single-source condition is shorter than that under the dual-source condition, and the running time increases with the increase of the number of array elements.

[0100] from Figure 7 It can be seen that after position error correction and mutual coupling error correction, the spatial-polarization domain joint search polarization generalized likelihood ratio test method and the minimum description length method still have a high success rate in estimating the number of sources under non-ideal conditions. Numerical simulations demonstrate the effectiveness of the position error correction algorithm and the mutual coupling error correction algorithm. After error correction, the number of sources can be effectively estimated. However, the proposed source number estimation method, the spatial-polarization domain joint search polarization generalized likelihood ratio test method, is derived under ideal conditions. Non-ideal factors will still have a certain impact on the performance of the proposed source number estimation method. To address this issue, the performance of the proposed algorithm under non-ideal conditions can be improved by combining it with manifold separation technology. Because the signal model of the proposed algorithm is applicable to wavefield models, it can be combined with manifold separation technology to improve its compatibility with unknown non-ideal models, but this will increase the computational complexity.

[0101] Table 1:

[0102]

[0103] Table 1 shows that when the number of sources is small, the search time of the spatial-polarization domain joint search polarization generalized likelihood ratio test method is shorter than the search time of the minimum description length method. This is because the spatial-polarization domain joint search polarization generalized likelihood ratio test method needs to judge the threshold each time to determine whether to perform the next search. If the value is not greater than the threshold in the next search, the method stops. When the number of sources is small, the spatial-polarization domain joint search polarization generalized likelihood ratio test method only needs to search the number of sources several times to determine the result, while the minimum description length method needs to search all possible results before making a judgment. Therefore, the spatial-polarization domain joint search polarization generalized likelihood ratio test method has a more obvious advantage over the minimum description length method when searching for a low number of sources.

Claims

1. A method for joint search of polarization generalized likelihood ratio test in spatial and polarization domains, characterized by: The estimation of the number of signal sources is modeled as a verification problem, with 0 and 1 representing the presence or absence of a signal. For the case of M>1 signal sources, M signal source verifications are required to complete the estimation. For the mth search, the DOA-polarization parameter needs to be known as The first m-1 incident signals can determine the DOA-polarization parameter: The existence or non-existence of the mth source of information, the hypothesis testing problem of the mth stage is modeled as: in Assume that there is a signal in the first m-1 stages, but no signal in the mth stage; Assume that there is a signal in the first m-1 stages and there is also a signal in the mth stage; the spatial domain and polarization domain joint search polarization generalized likelihood ratio test method includes the following steps: Step 1: Calculate the likelihood ratio L1(x), and then compare it with the predetermined threshold μ1. If the value is not greater than the threshold, stop searching and estimate the number of sources. Otherwise, update using the maximum likelihood estimation method Step 2: When m>1, A m-1 for where a m-1 For the The steering vector of the signal; Step 3: According to A m-1 Compute projection operator And calculate the likelihood ratio L m (x); Step 4: Calculate the L m (x) compared with the threshold value of the mth stage; Step 5: If L m (x) is less than the threshold, the search is terminated, and the estimated number of sources is Estimated parameters Step 6: If L m (x) is greater than the threshold, and all estimated incident angle parameters are updated using the maximum likelihood estimation method.

2. The method for joint search of polarization generalized likelihood ratio test in spatial and polarization domains according to claim 1, characterized in that: The likelihood ratio L1(x) described in step 1 is derived as follows: Assuming the noise variance is known, the likelihood ratio applicable to the scalar array is generalized to the polarized signal to obtain the likelihood ratio: Where μ m is the mth level threshold determined based on the mth level false alarm probability; a m is the estimated parameter The corresponding N×1-dimensional steering vector; K is the number of snapshots; is the projection operator: So for m=1, the likelihood ratio estimate is:

3. The method for joint spatial and polarization domain search polarization generalized likelihood ratio test according to claim 2, characterized in that: The threshold described in step 1 determines the success rate of source number estimation. Simplifying the likelihood ratio yields: Define the random variable z(t): Where z(t); t=1,...,K are K independent normally distributed complex random variables with mean μ m (t) The unit variable is, where μ m (t) is given by:

4. The method for joint spatial and polarization domain search polarization generalized likelihood ratio test according to claim 1, characterized in that: A described in step 2 m-1 It is the spatial-polarization array flow pattern of m-1 signals in M×(m-1) dimensions: in is a (m-1)×1 dimensional signal, s m (t) is the time domain waveform of the mth incident signal, and t is the sampling time.

5. The method for joint search of polarization generalized likelihood ratio test in spatial and polarization domains according to claim 1, characterized in that: The model used is the polarization conformal array signal model. The polarization spatial domain representation of the electromagnetic wave of the model is as follows: Consider a fully polarized electromagnetic wave incident signal incident on an array element. The incident direction of the electromagnetic wave signal can be determined only by the azimuth and elevation angles. Let θ be the azimuth angle, and θ is defined as the angle between the projection of the incident signal on the xoy plane and the positive direction of the x-axis. is the pitch angle, Defined as the angle between the incident signal and the positive direction of the z-axis; the polarization of the electromagnetic wave describes the trajectory of the electric field endpoint propagation in space, based on the two electric field components perpendicular to the propagation direction of the incident signal and The polarization state is determined by the amplitude ratio and phase difference of the polarization state, and γ is defined as the polarization auxiliary angle. and The amplitude ratio of the components; η is the polarization phase difference, η is defined as and The phase difference of the components is assumed to be a fully polarized wave incident from the narrowband far field. If the incident electromagnetic wave x(t) propagates in the direction of the wave, it can be expressed as: Where f0 is the carrier frequency; k = -(2π / λ) is the propagation wave number; is the propagation direction; s(t) is the complex baseband signal; E h =cosγ is the horizontal electric field strength, E v =sinγe jη is the electric field intensity in the vertical direction; the electric field intensity component of the electromagnetic wave is obtained by using the conversion relationship between the rectangular coordinate system and the spherical coordinate system, [E r ,E h ,E v ] T is the electric field intensity component in spherical coordinates, [E x ,E y ,E z ] T is the electric field intensity component in rectangular coordinates, E r If is 0, then: Similarly, the magnetic field intensity component can be obtained, and the electromagnetic field vector of the fully polarized wave in the rectangular coordinate system is:

6. The method for joint search of spatial and polarization domains for polarization generalized likelihood ratio testing according to claim 1, characterized in that: Applied to the polarization conformal array signal model, the traditional steering vector model is as follows: Because the curvature of the carrier causes conformal array elements to change their pattern gain during installation, traditional signal modeling based on consistent element pattern gain is no longer applicable. For the same incident signal, the azimuth and elevation angles in the local coordinate system of each element vary, resulting in different pattern gains for each element. Therefore, Euler rotation transformations must be used to solve for the steering vector when modeling the received signal of a conformal array. Consider an arbitrary conformal array with N elements, and the incident signal is a fully polarized wave in an arbitrary narrowband far field. The received signal data model is: Where x(t) is the N×1-dimensional received data vector at time t, where N is the number of array elements; s(t) is the M×1-dimensional incident signal vector, where M is the number of signal sources; and N(t) is the N×1-dimensional Gaussian white noise vector. is the spatial-polarization domain steering vector; In the formula is the spatial steering vector of the conformal array, is the polarization domain steering vector of the conformal array; ω=2π / λ, λ is the wavelength; r n is the position coordinate vector of the nth array element, and the specific expression is r n =[x n ,y n ,z n ];v m is the direction coordinate vector of the mth incident signal in the rectangular coordinate system, which is specifically expressed as Polarization response p mn =h m ·g n is the polarization vector h of the mth signal m In the nth array element pattern g n The projection on the global polar coordinate system is the signal polarization vector h m for: h m =h mh E h +h mv E v , tanγ=A h / A v ,η=φ h -f v ,(γ∈[0,π / 2],η∈[0,2π)), Where, E h is the horizontal basis vector of the electric field, E v is the basis vector in the vertical direction of the electric field, A h is the horizontal electric field amplitude, A v is the vertical electric field amplitude, φ h is the horizontal electric field phase, φ v is the vertical electric field phase.

Images