Coherent signal DOA estimation method based on space-time coding metasurface

Through the combination of spatiotemporal coding metasurface and l1-SVD algorithm, the problem of coherent signal processing in DOA estimation is solved, and high-precision and low-cost DOA estimation is achieved, which is suitable for complex environments.

CN118897250APending Publication Date: 2024-11-05NAT UNIV OF DEFENSE TECH
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202410938258.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-13
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

The prior art fails to effectively process coherent signals in DOA estimation, especially in complex propagation environments such as multipath reflection and human interference, which leads to mutual interfering and affects the estimation accuracy.

Method used

Using the coherent signal DOA estimation method based on the spatiotemporal coded metasurface, a coherent signal reception model is established by designing the spatiotemporal coded modulation sequence and fast Fourier transform theory, and the l1-SVD algorithm is used to perform coherent processing to reduce the influence of noise and realize the signal. accurate estimates.

Benefits of technology

This method preprocesses the signal through the spatiotemporal encoding metasurface, enriches signal amplitude and phase information, reduces hardware costs, reduces the number of receiving channels, improves the accuracy and robustness of DOA estimation, and is suitable for estimation of coherent and incoherent signals.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118897250A_ABST
    Figure CN118897250A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of metasurface and array signal processing, and particularly relates to a space-time coding metasurface-based coherent signal DOA (direction of arrival) estimation method, which comprises the following steps of: preprocessing coherent signals by using a space-time coding metasurface array through a coherent signal DOA estimation system consisting of the space-time coding metasurface array, an FPGA (field programmable gate array) control panel and a horn receiving antenna; rich signal amplitude and phase information is obtained, and signal processing of subsequent DOA estimation is facilitated; a # imgabs0 # algorithm is used for decoherence, so that the influence of noise is reduced, and the scale of a data matrix under the condition of multiple snapshots is reduced; the method is not only suitable for DOA estimation of coherent signals, but also effective for DOA estimation under incoherent conditions, and is more suitable for actual scenes.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of super surface and array signal processing, and in particular relates to a coherent signal DOA (Direction of Arrival) estimation method based on space-time coding super surface. Background Art

[0002] DOA estimation is one of the main research directions of array signal processing and has a wide range of applications in wireless sensing, communication, radar and other fields. The basic problem of DOA estimation is to determine the spatial position of multiple signals of interest that are simultaneously in a certain area of ​​space (the direction of arrival of each signal at the array, referred to as the direction of arrival). DOA estimation methods can be roughly divided into the following categories: the first category is beam scanning methods, such as the conventional beam forming (CBF) method and the minimum variance unbiased estimation (MVDR) algorithm; the second category is subspace-based methods, such as the multiple signal classification (MUSIC) algorithm and the subspace rotation invariance technique (ESPRIT) algorithm; the third category is fitting methods, such as the maximum likelihood method (ML); the fourth category is compressed sensing algorithms, such as the orthogonal matching pursuit (OMP) algorithm. Currently, the most commonly used method to achieve high resolution is the subspace-based algorithm, but this method often requires multiple sensors and corresponding receiving channels to receive signals. Therefore, achieving high-precision DOA estimation often requires complex calculations and expensive hardware.

[0003] In recent years, artificial electromagnetic metamaterials, especially metasurfaces, have achieved many revolutionary innovative research in the fields of radar imaging, wireless communication, electromagnetic stealth, etc. with their unprecedented performance. Digitally coded metasurfaces load diodes or bias circuits and introduce field programmable gate arrays (FPGAs) to switch the quantized phase coding distribution of surface array elements in real time to achieve all-round synchronous control of electromagnetic waves in the spatial, temporal, frequency and polarization domains, which greatly reduces the complexity, weight and cost of the system.

[0004] Due to its outstanding performance, metasurfaces have also been widely researched in the field of DOA estimation. The spatial and temporal coding of digitally coded metasurfaces is used to control the scattering field, so that the signal transmitted by the source is reflected by the digitally coded metasurface, passes through a receiving channel, and then the measurement matrix composed of the received signals is processed using a similar array signal processing method to achieve direction finding of the source. Finally, by introducing metasurfaces, the DOA estimation performance of a single-channel equivalent to a multi-channel array is achieved, balancing the contradiction between cost and DOA estimation accuracy.

[0005] However, in actual applications, due to complex propagation environments such as multipath reflection or human interference, the signals received by the metasurface antenna may be mutually coherent, including co-channel interference and multipath propagation signals formed by reflections from background objects. At present, although there are many studies on DOA estimation based on metasurfaces, few studies consider coherent signals. Summary of the invention

[0006] The purpose of the present invention is to propose a coherent signal DOA estimation method based on space-time coding metasurface, which utilizes the signal modulation capability of space-time coding metasurface to decoherent the preprocessed signal to obtain DOA estimation, and the method is easier to implement.

[0007] Specifically, the present invention comprises the following steps:

[0008] S1 designs the space-time coded modulation sequence and applies the fast Fourier transform theory to establish the coherent signal receiving model of the space-time coded metasurface DOA estimation system;

[0009] S1.1 Establishment of space-time coded metasurface DOA estimation system

[0010] This system consists of a space-time coding metasurface array, an FPGA control board and a horn receiving antenna; the space-time coding metasurface array consists of a sub-array of M×N space-time coding metasurface reflection units, M and N are both natural numbers greater than 1, and each space-time coding metasurface reflection unit is controlled by the FPGA control board to be in two coding states of 0 or 1, which have a modulation effect of 0° or 180° phase delay on the electromagnetic wave respectively; the horn receiving antenna is located above the space-time coding metasurface array, and is used to obtain the received signal, and the horn receiving antenna is connected to the spectrum analyzer to obtain the signal spectrum and extract the harmonic spectrum information.

[0011] The working state of the PIN diode carried by each space-time coding metasurface reflection unit can be independently switched by the FPGA control board in the form of digital coding. Because it can control the reflection unit to be in state 0 or 1 in the spatial domain, and can control the reflection unit to switch quickly between state 0 or 1 in the time domain, it is called a space-time coding metasurface reflection unit. The scattering field on its surface is regulated by controlling the space-time coding through the FPGA control board. Compared with the array antenna using multiple sensing channels, the present invention only needs one receiving channel to obtain the scattering field information, which reduces the number of receiving channels and reduces the hardware cost.

[0012] S1.2 Design of space-time coded modulation sequences

[0013] In order to establish the coherent signal reception model of the space-time coding metasurface array, a single-tone signal with the same frequency incident at multiple angles is taken as an example. When the coherent signal is incident at an angle θ i The light is irradiated on the space-time coding metasurface array and is captured by the horn receiving antenna after reflection. For one-dimensional angle estimation, each column of space-time coding metasurface reflection units can be regarded as a subarray, which in turn constitutes the entire space-time coding metasurface array. Therefore, the space-time coding metasurface model can be equivalent to the metasurface linear array model, and the model of the received signal can be expressed as:

[0014]

[0015] Among them, F c is the received signal frequency, A 0 and φ 0 are the initial amplitude and phase of the received signal, K is the wave number, D is the unit spacing of the linear array, n is the number of the linear array unit, n = 1, 2, ..., N, N is the total number of linear array units, i is the serial number of the coherent signal, i = 1, 2, ..., L, L is the total number of coherent signals, and t is the time variable of the received signal. n (t) is the space-time coded modulation sequence of the metasurface linear array model, expressed as:

[0016]

[0017] Among them, T p is the modulation period, τ n,ON and τ n,OFF are the moments when the encoding starts and ends in each modulation cycle, m is the number of modulation cycles, and Z represents a set of integers.

[0018] S1.3 Establishment of coherent signal reception model for space-time coded metasurface DOA estimation system

[0019] Space-time coded modulation sequence U of the metasurface linear array model n (t) Fourier expansion is:

[0020]

[0021] Among them, F p is the modulation frequency, which is related to the modulation period T p The reciprocal of each other, a n,k The k-order harmonic coefficient generated for the n-th linear array element is:

[0022]

[0023] Substituting equation (3) into equation (1), the coherent signal reception model of the space-time coded metasurface DOA estimation system is established as:

[0024]

[0025] The S2 transmitting horn antenna simulates the transmission of coherent signals. The signals are modulated by the space-time coding metasurface array and then received by the receiving horn antenna. A spectrum analyzer is used to obtain the harmonic spectrum of the coherent signals.

[0026] S2.1 This system is a passive signal receiving and processing system, so a transmitting horn antenna is required to simulate the external signal source incident during experimental testing. The horn antennas for transmitting and receiving coherent signals are both standard gain horn antennas. The placement of the transmitting horn antenna meets the far-field condition and can be approximately regarded as a plane wave at infinity. Use a signal source and a power divider to connect the two transmitting horn antennas to simulate the transmission of two co-frequency single-tone coherent signals. The receiving horn antenna is placed above the space-time coding metasurface array to receive the signal.

[0027] After the S2.2 coherent signal is modulated by the space-time coding metasurface array, the generated scattering field is received by the receiving horn antenna and connected to the spectrum analyzer for signal acquisition and sampling. The frequency domain signal obtained on the spectrum analyzer has the following characteristics: a frequency F c The fundamental spectral line, as well as the interval F p Positive integer multiples of the harmonic spectrum, where the amplitude values ​​of the fundamental and harmonic spectrum are expressed as the fundamental component coefficient γ 0 and the harmonic component coefficient γ k , combined with the mathematical relationship in the coherent signal reception model (5), γ 0 and γ k They are:

[0028]

[0029] S3 uses the mathematical relationship in the coherent signal reception model and the space-time coding matrix to recover the observation vector containing the coherent signal angle information.

[0030] Use a computer to read the harmonic spectrum of the spectrum analyzer, take the Q harmonics on the left and right sides of the fundamental wave, Q is the number of harmonics on each side, there are a total of 2Q+1 spectrum amplitude values, that is, k = -Q, -Q+1, ..., 0, ..., Q-1, Q, write the mathematical relations (6) and (7) in the coherent signal reception model in matrix form:

[0031]

[0032] It can be seen from equation (8) that the matrix on the left side of the equation is composed of the Fourier coefficients of the fundamental component and the harmonic component, which is defined as the space-time coding matrix. The value of each element in the space-time coding matrix is ​​determined by the space-time coded modulation sequence U of the metasurface linear array model. n (t) determines that the observation vector on the left side of the equation contains the angle information of the coherent signal, and the vector on the right side of the equation consists of the amplitude values ​​of the fundamental component and harmonic component of the spectrum obtained by the spectrum analyzer. The space-time coding matrix Ψ is expressed as:

[0033]

[0034] make

[0035]

[0036] γ=[γ -Q ,γ -Q+1 ,…,γ 0 ,…,γ Q-1 ,γ Q ] T (11)

[0037] Then, formula (8) is simplified to:

[0038]

[0039] The space-time coding matrix Ψ is a (2Q+1)×N matrix, and its generalized inverse matrix is ​​Ψ- 1 , then the observation vector containing the angle information of the coherent signal It can be expressed as:

[0040]

[0041] S4 uses compressed sensing algorithms 1 -After the SVD algorithm performs decorrelation processing, the DOA estimation result of the coherent signal is obtained;

[0042] S4.1 Observation vector containing the angle information of the coherent signal Sparse representation of

[0043] The observation vector to be solved is obtained through S3 The elements in this vector contain the angle information of multiple coherent signals. The angles contained in it are sparse in the spatial domain and can be obtained using compressed sensing algorithms. 1 -SVD algorithm for decorrelation:

[0044] Viewing observation vectors from a sparse perspective make

[0045]

[0046] Among them, the observation vector The size is N×1. N represents the total number of linear array units in the model, L represents the total number of coherent signals, and N>>L, for example, N=10L. Z is an overcomplete dictionary, which is a set of grids divided by the spatial domain. The size of Z is N×N θ , N θ Indicates the total number of grids divided into airspace, with a preset angle range S θ for

[0047] S θ ={-90°:1°:90°} (15)

[0048] So N θ =181; the size of the corresponding sparse signal vector x is N θ ×1, σ is the noise term;

[0049] Because the angle to be estimated is sparse in the spatial domain relative to the divided angle, most of the elements of the sparse signal vector x are 0, and the elements at the position of the sparse signal are non-zero. By solving the index position of the non-zero element, we can find the position of the sparse signal vector x in S θ The DOA of the signal is estimated in

[0050] Due to the influence of noise, some additional non-zero items will be generated. In the compressed sensing algorithm, the noise parameters are minimized and the l is minimized. 1 Norm processing, so the sparse signal vector x can be reconstructed by the objective equation in the following Lagrangian form:

[0051]

[0052] x is the reconstructed sparse vector, and λ is the regularization parameter used to balance x and The sparsity between ||*|| 2 Indicates l 2 norm,||*|| 1 Indicates l 1 norm;

[0053] S4.2 Dimensionality reduction of observation vectors The covariance matrix of

[0054] Calculate the observation vector The covariance matrix Y of is:

[0055]

[0056] Use singular value decomposition (SVD decomposition) to reduce the dimension of Y:

[0057] Y=UDV H (18)

[0058] Among them, U is N×N in size, V is T×T in size, both are unitary matrices, and D is N×T in size, a singular value matrix; define a selection matrix D K The size is N×K. It consists of a unit matrix of size K and a (TK)×K zero matrix. N and T represent the matrix dimensions after singular value decomposition. K represents the number of assumed coherent signals. K is generally less than N. The reduced-dimensional observation matrix Y is N×K in size. SV :

[0059] Y SV =YVD K =UDD K (19)

[0060] X SV Represents the covariance matrix of the sparse vector x after dimension reduction, σ SV The covariance matrix representing the noise after dimensionality reduction is:

[0061] X SV =xx H VD K (20)

[0062] σ SV =σσ H VD K (twenty one)

[0063] According to the sparse angle writing of formula (14), the observation matrix Y SV It can be expressed as

[0064] Y SV =ZX SV +σ SV ; (twenty two)

[0065] S4.3 Establishment based on l 1 -SVD algorithm’s decoherence objective function calculates the coherent signal’s incoming wave direction. According to the dimension reduction processing of S4.2, the objective equation of S4.1 becomes

[0066]

[0067] X SV becomes a matrix relative to x, and For X SV Each row vector l 2 The norm is used to replace each element in x, which makes the problem more efficient after transformation;

[0068] The objective equation problem is a second-order cone programming problem, and it is a convex problem. You can use MATLAB's convex toolbox cvx to solve the second-order cone programming problem. SV -ZX SV After vectorization, 2 Minimum sum of norms l 1 Under the two constraints of minimum norm, solve the X that satisfies the conditions. SV , and then find l for each row vector 2 norm, and obtain an N-dimensional column vector, where the index values ​​corresponding to the L non-zero values ​​are within the preset angle range S θ The angle in is the direction of the coherent signal.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] 1. Use the space-time coding metasurface to pre-process the coherent signal to obtain rich signal amplitude and phase information, which is beneficial to the subsequent signal processing of DOA estimation;

[0071] 2. This method uses l 1 -SVD algorithm is used for decorrelation, which reduces the influence of noise and the size of the data matrix in the case of multiple snapshots;

[0072] 3. This method is not only applicable to DOA estimation of coherent signals, but also effective for DOA estimation in incoherent cases, and is more suitable for practical scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 is the overall flow chart of the proposed method;

[0074] Figure 2 It is a schematic diagram of the structure of the coherent signal DOA estimation system based on the space-time coding metasurface;

[0075] Figure 3 It is a schematic diagram of the design of the space-time coded modulation sequence of the metasurface;

[0076] Figure 4 It is the spectrum preprocessing result diagram of the received coherent signal after FFT;

[0077] Figure 5 This is a physical picture of the space-time coding metasurface DOA prototype system used in the present invention;

[0078] Figure 6This is the result diagram of DOA estimation of coherent signal based on 1-norm singular value decomposition. DETAILED DESCRIPTION

[0079] The present invention is further described below with reference to the accompanying drawings and examples.

[0080] Figure 1 is an overall flow chart of the method.

[0081] The present invention provides a coherent signal DOA estimation method based on space-time coding metasurface, comprising the following steps:

[0082] S1 designs the space-time coded modulation sequence and applies the fast Fourier transform theory to establish the coherent signal receiving model of the space-time coded metasurface DOA estimation system;

[0083] The array scale of the space-time coding metasurface model used in this example is 20×20, with a total of 400 space-time coding metasurface reflection units. Each unit has two coding states, 0 or 1, which respectively represent a modulation effect of a phase delay of 0° or 180° on the electromagnetic waves irradiated to the metasurface. Each unit is efficiently controlled by the FPGA control board and can switch between the two states of 0 or 1 at high speed. Therefore, the designed 01 coding sequence can be input into the control board to control the metasurface.

[0084] S1.1 Establishment of space-time coded metasurface DOA estimation system

[0085] Establish as Figure 2 The coherent signal DOA estimation system based on space-time coding metasurface shown in the figure is composed of a space-time coding metasurface array, an FPGA control board and a horn receiving antenna; the space-time coding metasurface array is composed of M×N space-time coding metasurface reflection units as a subarray, M and N are both natural numbers greater than 1, and each space-time coding metasurface reflection unit is controlled by the FPGA control board to be in two coding states of 0 or 1, which has a modulation effect of 0° or 180° phase delay on the electromagnetic wave respectively; the horn receiving antenna is located above the space-time coding metasurface array, and is used to obtain the received signal, and the horn receiving antenna is connected to the spectrum analyzer to obtain the signal spectrum and extract the harmonic spectrum information.

[0086] S1.2 Design of space-time coded modulation sequences

[0087] Figure 3 This is a schematic diagram of the design of the space-time coding modulation sequence of the metasurface. Each column of the metasurface is equivalent to a subarray and is assigned the same coding sequence. The schematic diagram indicates that yellow is state 1 and blue is state 0, and the coding sequence is cyclic.

[0088] S1.3 Using Fast Fourier Transform Theory

[0089] S1.4 Establishment of coherent signal reception model for space-time coded metasurface DOA estimation system

[0090] The S2 transmitting horn antenna simulates the transmission of coherent signals. The signals are modulated by the space-time coding metasurface and then received by the receiving horn antenna. The harmonic spectrum of the coherent signals is obtained using a spectrum analyzer.

[0091] S2.1 This system is a passive signal receiving and processing system, so a transmitting horn antenna is required to simulate the external signal source incident during experimental testing. The horn antennas for transmitting and receiving coherent signals are both standard gain horn antennas. The placement of the transmitting horn antenna meets the far-field conditions and can be approximately regarded as a plane wave at infinity. A signal source and a power divider are used to connect the two transmitting horn antennas to simulate the transmission of two co-frequency single-tone coherent signals. The receiving horn antenna is placed above the space-time coding metasurface to receive the signal;

[0092] S2.2 obtains the harmonic spectrum, taking the amplitude values ​​of the fundamental wave and the Q-order harmonics to the left and right of the fundamental wave

[0093] Figure 4 It is the spectrum preprocessing result diagram of the received coherent signal after FFT, which is used to construct the observation vector

[0094] S3 uses the mathematical relationship in the coherent signal reception model and the space-time coding matrix to recover the observation vector containing the coherent signal angle information.

[0095] Get the observation vector containing the angle information of the coherent signal Matrix representation of

[0096] S4 uses compressed sensing algorithms 1 -After the SVD algorithm performs decorrelation processing, the DOA estimation result of the coherent signal is obtained.

[0097] S4.1 Observation vector containing the angle information of the coherent signal Sparse representation of

[0098] S4.2 Dimensionality reduction of observation vectors The covariance matrix of

[0099] S4.3 Establishment based on l 1 -SVD algorithm’s decoherence objective function calculates the direction of the coherent signal

[0100] Figure 5This is a physical picture of the space-time coding metasurface DOA prototype system used in the present invention. The system consists of a space-time coding metasurface array, an FPGA control board, and a horn receiving antenna: the space-time coding metasurface array consists of 20×20 space-time coding metasurface reflection units to form a subarray, and each space-time coding metasurface reflection unit is controlled by the FPGA control board to be in two coding states of 0 or 1, respectively, and has a modulation effect of 0° or 180° phase delay on the electromagnetic wave; the horn receiving antenna is located above the space-time coding metasurface array, used to obtain the received signal, and the horn receiving antenna is connected with a spectrum analyzer to obtain the signal spectrum and extract the harmonic spectrum information. Secondly, the placement of the transmitting horn antenna meets the far-field conditions to simulate the coherent signal transmission.

[0101] Figure 6 This is a diagram of the DOA estimation result of the coherent signal based on 1-norm singular value decomposition. The directions of the preset coherent signal source are -45° and 25°. It can be seen that the present invention can still obtain the DOA estimation result under the limitation of the coherent signal source, which provides a solution to the DOA estimation problem of coherent signals based on coded metasurface and has certain engineering application value.

Claims

1. A coherent signal DOA estimation method based on space-time coding metasurface, characterized in that: The method comprises the following steps: S1 designs the space-time coded modulation sequence and applies the fast Fourier transform theory to establish the coherent signal receiving model of the space-time coded metasurface DOA estimation system; S1.1 Establishment of space-time coded metasurface DOA estimation system The space-time coding metasurface DOA estimation system is composed of a space-time coding metasurface array, an FPGA control board and a horn receiving antenna; the space-time coding metasurface array is composed of M×N space-time coding metasurface reflection units to form a subarray, M and N are both natural numbers greater than 1, and each space-time coding metasurface reflection unit is controlled by the FPGA control board to be in two coding states of 0 or 1, which respectively have a modulation effect of 0° or 180° phase delay on the electromagnetic wave; the horn receiving antenna is located above the space-time coding metasurface array, and is used to obtain the received signal, and the horn receiving antenna is connected to the spectrum analyzer to obtain the signal spectrum and extract the harmonic spectrum information; S1.2 Design of space-time coded modulation sequences When the coherent signal has an incident angle θ i The light is irradiated on the space-time coding metasurface array and is captured by the horn receiving antenna after reflection. For one-dimensional angle estimation, each column of space-time coding metasurface reflection units can be regarded as a subarray, which in turn constitutes the entire space-time coding metasurface array. Therefore, the space-time coding metasurface model can be equivalent to the metasurface linear array model, and the model of the received signal can be expressed as: Among them, F c is the received signal frequency, A0 and φ0 are the initial amplitude and phase of the received signal, K is the wave number, D is the unit spacing of the linear array, n is the number of the linear array unit, n = 1, 2, ..., N, N is the total number of linear array units, i is the serial number of the coherent signal, i = 1, 2, ..., L, L is the total number of coherent signals, and t is the time variable of the received signal; U n (t) is the space-time coded modulation sequence of the metasurface linear array model, expressed as: Among them, T p is the modulation period, τ n,ON and τ n,OFF are the time when the code starts and ends in each modulation cycle, m is the number of modulation cycles, and Z represents a set of integers; S1.3 Establishment of coherent signal reception model for space-time coded metasurface DOA estimation system Space-time coded modulation sequence U of the metasurface linear array model n (t) Fourier expansion is: Among them, F p is the modulation frequency, which is related to the modulation period T p The reciprocal of each other, a n,k The k-order harmonic coefficient generated for the n-th linear array element is: Substituting equation (3) into equation (1), the coherent signal reception model of the space-time coded metasurface DOA estimation system is established as: The S2 transmitting horn antenna simulates the transmission of coherent signals. The signals are modulated by the space-time coding metasurface array and then received by the receiving horn antenna. A spectrum analyzer is used to obtain the harmonic spectrum of the coherent signals. The S2.1 space-time coded metasurface DOA estimation system is a passive signal receiving and processing system. Therefore, during experimental testing, a transmitting horn antenna is required to simulate the external signal source incident. The placement of the transmitting horn antenna meets the far-field condition and is regarded as a plane wave at infinity. A signal source and a power divider are used to connect the two transmitting horn antennas to simulate the transmission of two co-frequency single-tone coherent signals. The receiving horn antenna is placed above the space-time coded metasurface array to receive the signal. After the S2.2 coherent signal is modulated by the space-time coding metasurface array, the generated scattering field is received by the receiving horn antenna and connected to the spectrum analyzer for signal acquisition and sampling. The frequency domain signal obtained on the spectrum analyzer has the following characteristics: a frequency F c The fundamental spectral line, as well as the interval F p Positive integer multiples of the harmonic spectrum, where the amplitude values ​​of the fundamental and harmonic spectrum are expressed as the fundamental component coefficient γ0 and the harmonic component coefficient γ k , combined with the mathematical relationship in the coherent signal reception model (5), γ0 and γ k They are: S3 uses the mathematical relationship in the coherent signal reception model and the space-time coding matrix to recover the observation vector containing the coherent signal angle information; Use a computer to read the harmonic spectrum of the spectrum analyzer, take the Q harmonics on the left and right sides of the fundamental wave, Q is the number of harmonics on each side, there are a total of 2Q+1 spectrum amplitude values, that is, k = -Q, -Q+1, ..., 0, ..., Q-1, Q, write the mathematical relations (6) and (7) in the coherent signal reception model in matrix form: It can be seen from equation (8) that the matrix on the left side of the equation is composed of the Fourier coefficients of the fundamental component and the harmonic component, which is defined as the space-time coding matrix. The value of each element in the space-time coding matrix is ​​determined by the space-time coded modulation sequence U of the metasurface linear array model. n (t) determines that the observation vector on the left side of the equation contains the angle information of the coherent signal, and the vector on the right side of the equation is composed of the amplitude values ​​of the fundamental component and the harmonic component of the spectrum obtained by the spectrum analyzer; the space-time coding matrix Ψ is expressed as: make γ=[γ -Q ,c -Q+1 ,...,γ0,...,γ Q-1 ,c Q ] T (11) Then, formula (8) is simplified to: The space-time coding matrix Ψ is a (2Q+1)×N matrix, and its generalized inverse matrix is ​​Ψ -1 , then the observation vector containing the angle information of the coherent signal It can be expressed as: S4 uses the compressed sensing algorithm l1-SVD algorithm to perform decorrelation processing and obtain the DOA estimation result of the coherent signal; S4.1 Observation vector containing the angle information of the coherent signal Sparse representation of The observation vector to be solved is obtained through S3 The elements in this vector contain the angle information of multiple coherent signals. The angles contained in it are sparse in the spatial domain, and the compressed sensing algorithm is used for decorrelation: Viewing observation vectors from a sparse perspective make Among them, the observation vector The size is N×1. N represents the total number of linear array units in the model, L represents the total number of coherent signals, and N>>L is satisfied. Z is an overcomplete dictionary, which is a set of grids divided by the spatial domain. The size of Z is N×N θ , N θ Represents the total number of grids divided in the spatial domain, and the size of the corresponding sparse signal vector x is N θ ×1, σ is the noise term; Because the angle to be estimated is sparse in the spatial domain relative to the divided angle, most of the elements of the sparse signal vector x are 0, and the elements at the position of the sparse signal are non-zero. By solving the index position of the non-zero element, we can find the position of the sparse signal vector x in S θ The DOA of the signal is estimated in Due to the influence of noise, some additional non-zero terms will be generated. In the compressed sensing algorithm, the noise parameter is minimized and the l1 norm is minimized, so the sparse signal vector x can be reconstructed by the objective equation in the following Lagrangian form: x is the reconstructed sparse vector, and λ is the regularization parameter used to balance x and The sparsity between them, ||*||2 represents the l2 norm, and ||*||1 represents the l1 norm; S4.2 Dimensionality reduction of observation vectors The covariance matrix of Calculate the observation vector The covariance matrix Y of is: Use singular value decomposition to reduce the dimension of Y: Y=UDV H (17) Among them, U is N×N in size, V is T×T in size, both are unitary matrices, and D is N×T in size, a singular value matrix; define a selection matrix D K The size is N×K. It consists of a unit matrix of size K and a (TK)×K zero matrix. N and T represent the matrix dimensions after singular value decomposition. K represents the number of assumed coherent signals. K is generally less than N. The reduced-dimensional observation matrix Y is N×K in size. SV : Y SV =YVD K =USD K (18) X SV Represents the covariance matrix of the sparse vector x after dimension reduction, σ SV The covariance matrix representing the noise after dimensionality reduction is: X SV =xx H VD K (19) s SV =ss H VD K (20) According to the sparse angle writing of formula (14), the observation matrix Y SV It can be expressed as Y SV =ZX SV +s SV (21) S4.3 establishes the decoherence objective function based on the l1-SVD algorithm and calculates the direction of the coherent signal. According to the dimensionality reduction processing in S4.2, the objective equation of S4.1 becomes X SV becomes a matrix relative to x, and For X SV The l2 norm of each row vector is used to replace each element in x, which makes the problem more efficient. The objective equation problem is a second-order cone programming problem, and it is a convex problem. The convex toolbox cvx of MATLAB is used to solve the second-order cone programming problem. SV -ZX SV Vectorized L2 norm minimum sum Under the two constraints of the minimum l1 norm, we can solve X that satisfies the conditions. SV , and then calculate the l2 norm of each row vector to obtain an N-dimensional column vector, where the index values ​​corresponding to the L non-zero values ​​are within the preset angle range S θ The angle in is the direction of the coherent signal.

2. A coherent signal DOA estimation method based on space-time coding metasurface according to claim 1, characterized in that: In S1.2, a coherent signal reception model of a space-time coded metasurface array is established by taking multiple single-tone signals of the same frequency incident at different angles as an example.

3. A coherent signal DOA estimation method based on space-time coding metasurface according to claim 1, characterized in that: In S2.1, the horn antennas for transmitting and receiving coherent signals are both standard gain horn antennas.

4. A coherent signal DOA estimation method based on space-time coding metasurface according to claim 1, characterized in that: In S4.1, the compressed sensing algorithm used is the l1-SVD algorithm.

5. A coherent signal DOA estimation method based on space-time coding metasurface according to claim 1, characterized in that: In S4.1, the preset angle range S θ For S θ = {-90°:1°:90°}, the total number of grids N divided into airspace θ =181.

Citation Information

Cited By

  • Fundamental wave and multi-order harmonic combined regulation and control method based on space-time coding metasurface and multi-source direction finding method

    CN119471559A

  • Space-time modulation metasurface DOA estimation method based on unit non-uniform arrangement

    CN119556230A

  • Sensing and beam manipulation decoupling method based on space-time coding metamaterial

    CN119764855A

  • Unit non-uniform metasurface DOA estimation method and device based on depth expansion

    CN119805363A