A satellite navigation anti-interference polarization antenna array reconstruction method

Abstract

The present invention provides a method for reconstructing a satellite navigation anti-interference polarized antenna array. The present invention selects only a portion of radio frequency channels to access the antenna array and transmit signals. A correlation measurement reduction algorithm is used to quickly and accurately obtain a tri-polarized antenna array structure configuration that can output the maximum CNR when different numbers of radio frequency channels are selected. The position of the selected radio frequency channel is determined, and by controlling the on / off state of the radio frequency switch between the antenna and the radio frequency channel, the selected radio frequency channel is connected to the corresponding antenna for subsequent adaptive signal processing. At the same time, the radio frequency switch corresponding to the unselected radio frequency channel is disconnected and connected to a matching load to complete the reconstruction of the tri-polarized antenna array. Based on the radio frequency channel selection and reconstruction strategy, the present invention provides a fast and efficient tri-polarized antenna array reconstruction method. After the number of radio frequency channels used for array reconstruction is determined, the reconstructed tri-polarized antenna array is used to ensure the maximum CNR output.

Description

Technical Field

[0001] The present invention relates to the field of array reconstruction, and in particular to an antenna array reconstruction method. Background Art

[0002] Satellite navigation technology is widely used in various fields. However, the satellite signal power transmitted to ground receiver terminals is only -133dB, which is very weak and highly susceptible to interference. Although spread spectrum technology provides some anti-interference capabilities, it is still difficult to resist high-intensity human interference. To ensure that receivers can still perform positioning and calculations in strong electromagnetic interference environments, adaptive anti-interference technology based on antenna arrays has flourished in recent years.

[0003] Currently, antenna arrays available for satellite navigation anti-interference systems are primarily categorized into two types based on antenna type: scalar and vector. Scalar array signal processing relies solely on spatial phase information differences for filtering. When interference and signal directions are close or identical, anti-interference capabilities decline sharply or even disappear. Vector antenna arrays, on the other hand, offer diversified element polarization selectivity, enabling them to distinguish the polarization characteristics of received signals, improving anti-interference performance. Currently, the most widely studied vector array is the tri-polarized antenna array, which consists of multiple orthogonal dipole antennas (three mutually perpendicular dipoles). Compared to a scalar array with the same layout and number of antennas, a tri-polarized antenna array offers enhanced anti-interference capabilities and can still deliver a high carrier-to-noise ratio (CNR) in the face of co-directional interference. However, the array also uses significantly more RF channels than a scalar array, increasing both hardware design costs and the computational burden of adaptive signal processing. To solve this problem, array reconstruction technology has been widely studied in recent years. This technology is dedicated to selecting only some antennas and RF channels to receive signals, reconfiguring the array, reducing design hardware and software costs, while ensuring maximum protection of anti-interference performance.

[0004] Current array reconfiguration techniques typically use mathematical optimization methods or genetic algorithms to optimize element placement. This process is cumbersome, computationally complex, slow, and inefficient, making it difficult to output array reconfiguration results in real time. To more efficiently obtain the optimal configuration for array reconfiguration, this paper proposes a method for reconfiguring a satellite navigation anti-interference polarized antenna array. By using radio frequency switches to select the use of certain radio frequency channels, this method enables rapid reconfiguration of a tri-polarized antenna array while maintaining high anti-interference performance. Summary of the Invention

[0005] To overcome the shortcomings of the existing technology, the present invention provides a method for reconfiguring a satellite navigation anti-interference polarized antenna array. This method selects only a portion of RF channels to access the antenna array and transmit signals. Using a correlation measurement reduction algorithm, the method quickly and accurately determines the tri-polarized antenna array structure configuration that can output the maximum CNR when different numbers of RF channels are selected, thereby determining the position of the selected RF channel. By controlling the on / off state of the RF switch between the antenna and the RF channel, the selected RF channel is connected to the corresponding antenna for subsequent adaptive signal processing. Simultaneously, the RF switch corresponding to the unselected RF channel is disconnected and connected to a matching load, completing the reconstruction of the tri-polarized antenna array.

[0006] The technical solution adopted by the present invention to solve the technical problem includes the following steps:

[0007] Step a: Obtain the antenna array element position coordinate matrix: The tri-polarized antenna array consists of N three orthogonal electric dipole antennas. The antenna array element position coordinates are:

[0008]

[0009] where p l =[x l ,y l ,z l ] represents the coordinates of the lth three orthogonal electric dipoles, l = 1, ... N;

[0010] Step b: Obtain the spatial direction and polarization characteristics of the signal and interference: In the current scenario, there is 1 signal and M interferences. Indicates the elevation and azimuth angles of the signal, represents the elevation angle and azimuth angle of the mth interference, m=1,...M, (γ0,η0) represents the polarization auxiliary angle and polarization phase difference of the signal, (γ i ,η i ) represents the polarization auxiliary angle and polarization phase difference of the i-th interference, and the angle of arrival (DOA) of the signal and interference is calculated as:

[0011]

[0012] Step c: Calculate the polarization spatial joint steering vector: Combining equations (1) and (2), calculate the spatial steering vectors of the signal and interference:

[0013]

[0014] Where λ is the wavelength of the signal;

[0015] The polarization domain steering vectors of the signal and interference are calculated as:

[0016]

[0017] Combining equations (3) and (4), the polarization spatial joint steering vector of the signal and interference is calculated as:

[0018]

[0019] in, Represents the Kronecker product of matrices.

[0020] Step d: Calculate the array manifold matrix: Calculate s according to formula (5) pi , calculate the array manifold matrix U of the signal plus interference X and the interference array manifold matrix U I ,

[0021] U X =[s p0 ,s p1 ,…,s pM ] (6)

[0022] U I =[s p1 ,s p2 ,…,s pM ] (7)

[0023] Step e: Establish the array reconstruction problem objective function and constraints: Define the RF channel selection vector x∈R 3N × 1 , the constituent element x d Either 0 or 1, d = 1, ... 3N, 0 means that the RF antenna is not used, and 1 means that the RF channel is selected; based on the minimum variance distortionless response (MVDR) algorithm, combined with formula (6) and formula (7), the optimal CNR expression of the tri-polarized antenna array output is:

[0024]

[0025] Among them, P s is the power of the satellite signal after demodulation with the pseudo-random noise (PRN) code, P n is the power spectral density of the noise, g n Represents the noise gain, det() represents the determinant of the matrix, and diag(x) represents taking all elements of vector x to construct a diagonal matrix;

[0026] Select K RF channels to reconstruct the tri-polarized antenna array, and the goal is to obtain the maximum CNR. Establish the objective function and constraints of the array reconstruction problem:

[0027]

[0028] stx d (xd -1)=0d=i...3N

[0029] 1 T x=K (9)

[0030] Among them, 1 represents a 3N×1 dimensional vector whose elements are all 1;

[0031] Step f: Obtain a reconstructed tri-polarized antenna array containing K radio frequency channels and capable of outputting the maximum CNR: Based on the obtained objective function F(x), the positions of the K radio frequencies of the reconstructed tri-polarized antenna array capable of outputting the maximum CNR are selected using the correlation measurement subtraction method;

[0032] After the correlation measurement deletion method is executed, x contains K 1s and 3N-K 0s. A value of 1 indicates that the corresponding RF channel is selected, and a value of 0 indicates that the corresponding RF channel is not selected. The K selected RF channels are connected to the antennas at the corresponding positions through the RF switches to form a new reconstructed tri-polarized antenna array.

[0033] Step g: Calculation of adaptive anti-interference weights of the reconstructed tri-polarized antenna array: Using the MVDR algorithm, the anti-interference weights of the reconstructed tri-polarized antenna array are calculated as follows:

[0034]

[0035] Among them, R N is the channel interference plus noise statistical covariance matrix;

[0036] Step h: Adaptive beamforming of the reconstructed tri-polarized antenna array: The anti-interference weight calculated by formula (10) is multiplied by the channel data for weighted processing, and the array output is:

[0037]

[0038] Wherein, Y(t) is the data transmitted by the K radio frequency channels selected after reconstruction.

[0039] The specific steps of the related measurement subtraction method in step f are as follows:

[0040] Step f.1) Initialize x = 1, select all RF channels, and iterate n = 1;

[0041] Step f.2) Determine x d ≠0 is true, if true, let x d = 0, calculate f(i) = F(x), and then let x d =1, if x d ≠0 does not hold, let f(i)=0;

[0042] Step f.3) Calculate j = argmaxd＝1，...，3N f(d), where arg represents the value of d when f(d) takes its maximum value. Let x j =0, n plus 1;

[0043] Step f.4) determines whether n=3N-K+1 holds. If so, the algorithm ends and outputs the current x. If n=3N-K+1 does not hold, return to step f.2) and continue execution.

[0044] The present invention provides a fast and efficient method for reconfiguring a tri-polarized antenna array based on a radio frequency channel selection and reconstruction strategy. After specifying the number of radio frequency channels used for array reconstruction, the tri-polarized antenna array reconstructed using the present method can ensure the maximum output CNR. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of the method implementation of the present invention.

[0046] Figure 2 This is the polarization sensitive array model used in the present invention.

[0047] Figure 3 Schematic diagram of RF channel selection for reconstructing a tri-polarized antenna array when N=9 and K=18.

[0048] Figure 4 When N=9 and K=18, the beam pattern of the tri-polarized antenna array is reconstructed.

[0049] Figure 5 When N = 9, the maximum CNR is obtained by the correlation measurement deletion method and the enumeration method under different numbers of RF channels. DETAILED DESCRIPTION

[0050] The present invention will be further described below with reference to the accompanying drawings and examples.

[0051] Step a: Obtain the antenna array element position coordinate matrix: The tri-polarized antenna array consists of N three orthogonal electric dipole antennas, and the antenna array element position coordinates are:

[0052]

[0053] where p l =[x l ,y l ,z l ] represents the coordinates of the lth three orthogonal electric dipoles, l = 1, ... N;

[0054] Step b: Obtain the spatial direction and polarization characteristics of the signal and interference: In the current scenario, there is 1 signal and M interferences. Indicates the elevation and azimuth angles of the signal, represents the elevation angle and azimuth angle of the mth interference, m=1,...M, (γ0,η0) represents the polarization auxiliary angle and polarization phase difference of the signal, (γ i ,η i ) represents the polarization auxiliary angle and polarization phase difference of the i-th interference, and the angle of arrival (DOA) of the signal and interference is calculated as:

[0055]

[0056] Step c: Calculate the polarization spatial joint steering vector: Combining equations (1) and (2), calculate the spatial steering vectors of the signal and interference:

[0057]

[0058] Where λ is the wavelength of the signal;

[0059] The polarization domain steering vectors of the signal and interference are calculated as:

[0060]

[0061] Combining equations (3) and (4), the polarization spatial joint steering vector of the signal and interference is calculated as:

[0062]

[0063] in, Represents the Kronecker product of matrices.

[0064] Step d: Calculate the array manifold matrix: Calculate s according to formula (5) pi , calculate the array manifold matrix U of the signal plus interference X and the interference array manifold matrix U I ,

[0065] U X =[s p0 ,s p1 ,…,s pM ] (6)

[0066] U I =[s p1 ,s p2 ,…,s pM ] (7)

[0067] Step e: Establish the array reconstruction problem objective function and constraints: Define the RF channel selection vector x∈R 3N×1 , the constituent element x dEither 0 or 1, d = 1, ... 3N, 0 means that the RF antenna is not used, and 1 means that the RF channel is selected; based on the minimum variance distortionless response (MVDR) algorithm, combined with formula (6) and formula (7), the optimal CNR expression of the tri-polarized antenna array output is:

[0068]

[0069] Among them, P s is the power of the satellite signal after demodulation with the pseudo-random noise (PRN) code, P n is the power spectral density of the noise, g n Represents the noise gain, det() represents the determinant of the matrix, and diag(x) represents taking all elements of vector x to construct a diagonal matrix;

[0070] Select K RF channels to reconstruct the tri-polarized antenna array, and the goal is to obtain the maximum CNR. Establish the objective function and constraints of the array reconstruction problem:

[0071]

[0072] stx d (x d -1)=0d=1...3N

[0073] 1 T x=K (9)

[0074] Among them, 1 represents a 3N×1 dimensional vector whose elements are all 1;

[0075] Step f: Obtain a reconstructed tri-polarized antenna array containing K radio frequency channels and capable of outputting the maximum CNR: Based on the obtained objective function F(x), a correlation measurement subtraction method is used to select the positions of the K radio frequency channels of the reconstructed tri-polarized antenna array that can output the maximum CNR. The specific steps of the correlation measurement subtraction algorithm are as follows:

[0076] Step f.1) Initialize x = 1, select all RF channels, and iterate n = 1;

[0077] Step f.2) Determine x d ≠0 is true, if true, let x d = 0, calculate f(i) = F(x), and then let x d =1, if x d ≠0 does not hold, let f(i)=0;

[0078] Step f.3) Calculate j = argmax d＝1，...，3N f(d), where arg represents the value of d when f(d) takes its maximum value. Let x j =0, n plus 1;

[0079] Step f.4) Determine whether n = 3N - K + 1. If so, the algorithm ends and outputs the current x. If not, return to step f.2) and continue execution.

[0080] After the correlation measurement deletion method is executed, x contains K 1s and 3N-K 0s. A value of 1 indicates that the corresponding RF channel is selected, and a value of 0 indicates that the corresponding RF channel is not selected. The K selected RF channels are connected to the antennas at the corresponding positions through the RF switches to form a new reconstructed tri-polarized antenna array.

[0081] Step g: Calculation of adaptive anti-interference weights of the reconstructed tri-polarized antenna array: Using the MVDR algorithm, the anti-interference weights of the reconstructed tri-polarized antenna array are calculated as follows:

[0082]

[0083] Among them, R N is the channel interference plus noise statistical covariance matrix;

[0084] Step h: Adaptive beamforming of the reconstructed tri-polarized antenna array: The anti-interference weight calculated by formula (10) is multiplied by the channel data for weighted processing, and the array output is:

[0085]

[0086] Wherein, Y(t) is the data transmitted by the K radio frequency channels selected after reconstruction.

[0087] The present invention proposes a reconstruction method of a tri-polarized antenna array for satellite navigation anti-interference, and the specific implementation process is as follows: Figure 1 As shown below. Figure 2 The tri-polarized antenna array shown in the figure demonstrates the GPS satellite signal reception, anti-interference, and array reconstruction process. Nine tri-polarized antennas are evenly arranged in a linear array with a signal half-wavelength spacing. The initial array contains 27 RF channels. The goal of this example is to construct a new array with only 18 RF channels and ensure that the array can output the maximum CNR. There is one signal and three interferences incident on the array. The satellite signal power P s =-160dBW, noise power spectral density P n =-204dBW-Hz, noise gain g n =1, and other parameter settings of signal and interference are shown in Table 1:

[0088] Table 1 Signal and interference parameters

[0089]

[0090] The specific steps for polarization-sensitive array reconstruction are as follows:

[0091] Step 1: Calculate the polarization domain spatial steering vector s of the signal and interference using formulas (1) to (5) pi , i=0,1,2,3, and according to formula (6) and formula (7), the array manifold matrix U of signal plus interference is obtained X and the interference array manifold matrix U I .

[0092] Step 2: Based on the U obtained in step 1 X and U I , combined with formula (8) and formula (9), the objective function and constraints of the reconstruction problem are calculated:

[0093]

[0094] stx d (x d -1)=0d=1...2

[0095] 1 T x=18 (9)

[0096] Step 3: Based on F(x) obtained in step 2, use the relevant measurement subtraction method steps in Table 2 to solve the result of selecting the 18 RF channels that make up the optimal array.

[0097] Table 2 Related measurement deletion method

[0098]

[0099] Figure 3 The results of the selected use of the radio frequency channels of the reconstructed tri-polarized antenna array are given, where the red cross indicates that the radio frequency channel is not used, and the blue circle indicates that the radio frequency channel is used. Figure 3 , connect the RF channels corresponding to the 18 blue circles to the antenna, and disconnect the RF channels corresponding to the 9 red crosses from the antenna to form the optimal reconstructed three-polarization antenna array.

[0100] Step 4: Use the reconstructed tri-polarized antenna array obtained in step 3 to receive signals, calculate new adaptive anti-interference weights and perform weighted processing on the channel data according to formulas (10) and (11). Figure 4 The beam pattern of the reconstructed tri-polarized array containing 18 RF channels is given.

[0101] Figure 5 The results of the CNR output after array reconstruction using the correlation measurement deletion method and the true maximum CNR output after array reconstruction using the enumeration method are given when different numbers of RF channels are selected. Figure 5It can be seen that the maximum CNR obtained by the two methods is basically the same. Among them, when K=18, the maximum CNR obtained by the correlation measurement deletion method and the enumeration method are 51.92dB and 51.95dB respectively, which proves the effectiveness of the correlation measurement deletion method in reconstructing the array. Moreover, for the number of RF channels K=18, the enumeration method should be compared The array output results of various RF channel combinations require calculating the objective function F(x) 4,686,825 times to obtain the optimal RF channel configuration that maximizes the CNR. The correlation measurement reduction method, on the other hand, only requires calculating F(x) 27 × (27 - 18) = 243 times, a computationally intensive approach that is far less complex than the enumeration method. This allows for faster and more efficient acquisition of the optimally configured reconstructed tri-polarized antenna array.

Claims

1. A satellite navigation anti-interference polarization antenna array reconstruction method, characterized in that The steps include: Step a: Obtain the antenna array element position coordinate matrix: The tri-polarized antenna array consists of N three orthogonal electric dipole antennas. The antenna array element position coordinates are: where p l =[x l ,y l ,z l ] represents the coordinates of the lth three orthogonal electric dipoles, l = 1, ... N; Step b: Obtain the spatial direction and polarization characteristics of the signal and interference: In the current scenario, there is 1 signal and M interferences. Indicates the elevation and azimuth angles of the signal, represents the elevation angle and azimuth angle of the mth interference, m=1,...M, (γ0,η0) represents the polarization auxiliary angle and polarization phase difference of the signal, (γ i ,η i ) represents the polarization auxiliary angle and polarization phase difference of the i-th interference, and the angle of arrival (DOA) of the signal and interference is calculated as: Step c: Calculate the polarization spatial joint steering vector: Combining equations (1) and (2), calculate the spatial steering vectors of the signal and interference: Where λ is the wavelength of the signal; The polarization domain steering vectors of the signal and interference are calculated as: Combining equations (3) and (4), the polarization spatial joint steering vector of the signal and interference is calculated as: in, represents the Kronecker product of the matrix; Step d: Calculate the array manifold matrix: Calculate s according to formula (5) pi , calculate the array manifold matrix U of the signal plus interference X and the interference array manifold matrix U I , U X =[s p0 ,s p1 ,…,s pM ] (6) U I =[s p1 ,s p2 ,…,s pM ] (7) Step e: Establish the array reconstruction problem objective function and constraints: Define the RF channel selection vector x∈R 3N×1 , the constituent element x d Either 0 or 1, d = 1, ... 3N, 0 means that the RF antenna is not used, and 1 means that the RF channel is selected; based on the minimum variance distortionless response (MVDR) algorithm, combined with formula (6) and formula (7), the optimal CNR expression of the tri-polarized antenna array output is: Among them, P s is the power of the satellite signal after demodulation with the pseudo-random noise (PRN) code, P n is the power spectral density of the noise, g n Represents the noise gain, det() represents the determinant of the matrix, and diag(x) represents taking all elements of vector x to construct a diagonal matrix; Select K RF channels to reconstruct the tri-polarized antenna array, and the goal is to obtain the maximum CNR. Establish the objective function and constraints of the array reconstruction problem: s.t.x d (x d -1)=0 d=1...3N 1 T x=K (9) Among them, 1 represents a 3N×1 dimensional vector whose elements are all 1; Step f: Obtain a reconstructed tri-polarized antenna array containing K radio frequency channels and capable of outputting the maximum CNR: Based on the obtained objective function F(x), the positions of the K radio frequencies of the reconstructed tri-polarized antenna array capable of outputting the maximum CNR are selected using the correlation measurement subtraction method; After the correlation measurement deletion method is executed, x contains K 1s and 3N-K 0s. A value of 1 indicates that the corresponding RF channel is selected, and a value of 0 indicates that the corresponding RF channel is not selected. The K selected RF channels are connected to the antennas at the corresponding positions through the RF switches to form a new reconstructed tri-polarized antenna array. The specific steps of the correlation measurement subtraction method are as follows: Step f.1) Initialize x = 1, select all RF channels, and iterate n = 1; Step f.2) Determine x d ≠0 is true, if true, let x d = 0, calculate f(i) = F(x), and then let x d =1, if x d ≠0 does not hold, let f(i)=0; Step f.3) Calculate j = argmax d＝1，...，3 Nf(d), where arg represents the value of d when f(d) takes its maximum value. Let x j =0, n plus 1; Step f.4) Determine whether n = 3N - K + 1. If so, the algorithm ends and outputs the current x. If not, return to step f.2) and continue execution. Step g: Calculation of adaptive anti-interference weights of the reconstructed tri-polarized antenna array: Using the MVDR algorithm, the anti-interference weights of the reconstructed tri-polarized antenna array are calculated as follows: Among them, R N is the channel interference plus noise statistical covariance matrix; Step h: Adaptive beamforming of the reconstructed tri-polarized antenna array: The anti-interference weight calculated by formula (10) is multiplied by the channel data for weighted processing, and the array output is: Wherein, Y(t) is the data transmitted by the K radio frequency channels selected after reconstruction.

Images