CN118655518A - Signal Direction of Arrival Estimation Method Based on Constructing Received Data Matrix - Google Patents

Abstract

The invention discloses a signal direction of arrival estimation method based on a construction of a received data matrix, which comprises the following steps: step 1, giving a continuous pulse echo quantity condition which can be used for coherent accumulation processing; step 2, demodulating echo signals received by each antenna array element to obtain intermediate frequency signals: step 3, carrying out beam forming processing after digital sampling; step 4, pulse compression processing is carried out to obtain a pulse compression processing result; step 5, continuously processing a plurality of echo signals to obtain a data matrix, and performing Fourier transformation in a slow time domain to obtain a moving target RD plane; step 6, detecting a moving target and determining the position of a target point; step 7, performing pulse compression processing on intermediate frequency signals obtained by continuous echoes acquired by each array element, and constructing a received data matrix; and 8, processing the received data matrix to obtain a direction-of-arrival estimated value. The method effectively reduces the influence of clutter and noise on the estimation performance of the direction of arrival.

Description

Signal Direction of Arrival Estimation Method Based on Constructing Received Data Matrix

Technical Field

The invention belongs to the technical field of radar antenna array signal processing, and in particular relates to a signal arrival direction estimation method based on constructing a receiving data matrix.

Background Art

Regarding the problem of DOA estimation of LFM (linear frequency modulation) signals commonly used in radar, since LFM signals are non-stationary signals, and traditional spatial spectrum estimation algorithms such as ESPRIT and MUSIC are only adaptable to stationary signals under high signal-to-noise ratio conditions, and also require the number of signals contained in the received data to be known, some scholars have proposed a method to achieve DOA estimation through time-frequency transform, such as fractional Fourier transform and Radon-Wigner transform, but there are problems with high algorithm complexity and insufficient performance under low signal-to-noise ratio conditions.

Summary of the invention

The object of the present invention is to provide a signal direction of arrival estimation method based on constructing a received data matrix to solve the problem of direction of arrival estimation of LFM signals existing in the above-mentioned prior art, especially the problem of direction of arrival estimation of target echo signals under low signal-to-noise ratio conditions.

In order to achieve the above object, the present invention adopts the following technical solutions:

A method for estimating the direction of arrival of a signal based on constructing a received data matrix specifically comprises the following steps:

Step 1: The number M of continuous pulse echoes that can be used for coherent accumulation processing by DAR must satisfy:

Where:

M—the number of consecutive pulse echoes that can be used for coherent accumulation processing;

c—speed of light;

B—Bandwidth of the LFM signal;

v— radial velocity of the target;

_Tr — the pulse repetition interval of the LFM signal transmitted by the radar;

Step 2: Demodulate the echo signal received by each antenna element of the DAR to obtain an intermediate frequency signal s _IF,n (t,t _m ):

Where:

s _IF,n (t,t _m )—intermediate frequency signal;

τ _n,m — round trip delay;

t—time;

T—pulse width;

β—backscatter coefficient of point target;

exp(·) — exponential function;

j—imaginary unit;

K—frequency modulation slope of LFM signal, K=B/T;

_f0 —transmitted signal carrier frequency;

r ₀ — initial distance of the target;

m—the sequence number of the pulse signal, m∈[1,M];

θ—target angle;

n—the serial number of the antenna array element of DAR, n∈[0,N-1], N is the number of antenna array elements;

d—antenna array element spacing;

Step 3: After digital sampling of the intermediate frequency signal s _IF,n (t,t _m ), a beamforming process is performed as shown in the following formula to obtain a beamforming output result, namely:

Where:

W—beamforming weight vector;

H—conjugate transpose;

Step 4: Output the beamforming results Perform pulse compression processing to obtain the pulse compression processing result After processing M echo signals continuously, the processed data are arranged in parallel to obtain a data matrix

Step 5, performing Fourier transform processing on the data matrix obtained in step 4 in the slow time _tm domain to obtain the RD plane of the moving target;

Step 6, completing the moving target detection on the RD plane obtained in step 5, thereby determining the distance position n _s where the moving target is located;

Step 7: demodulate and pulse compress the M echo signals received by the N array elements, collect the _nth complex values of each echo pulse compression result of each array element, and construct an N×M dimensional receiving data matrix X for DOA estimation, that is:

Step 8: For the constructed receiving data matrix X, it is known that it contains only one target echo signal, and multiple signal classification method or signal parameter estimation algorithm based on rotation invariance technology is used for processing to obtain the estimated value of the arrival direction of the target echo.

Furthermore, in step 6, the moving target detection adopts a constant false alarm detection algorithm.

Furthermore, in step 8, the signal parameter estimation algorithm is a multiple signal classification method or a rotation-invariant technology.

Compared with the prior art, the technical effects of the present invention are as follows:

1. The present invention uses the RD plane target detection results to determine the target position, and then extracts the data of the corresponding distance point from the pulse compression results, effectively reducing the impact of clutter and noise on the direction of arrival estimation performance;

2. The present invention constructs a new receiving data matrix by continuously extracting pulse compression data of target position points in multiple echoes, thereby enhancing the DOA estimation performance under low signal-to-noise ratio conditions;

3. The present invention constructs a receiving data matrix for each target echo, that is, it has been clarified that there is one and only one arrival signal in the newly constructed receiving data matrix, which creates favorable conditions for the subsequent application of MUSIC or ESPRIT algorithm.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is the RD plane;

Figure 2 shows the spatial spectrum estimation obtained by the MUSIC algorithm.

The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments.

DETAILED DESCRIPTION

The technical terms involved in the present invention are:

(1) DAR: Digital Array Radar, which can detect targets through the range-Doppler plane and estimate the distance and speed parameters of the target.

(2) RD plane: Range-Doppler plane.

(3)DOA: Direction of Arrival.

(4) LFM signal: Linear Frequency Modulation signal.

(5) PRI: Pulse Repetition Interval, which refers to the pulse repetition interval of the radar signal.

(6)MUSIC: Multiple Signal Classification.

(7)ESPRIT: Estimating Signal Parameter via Rotational Invariance Techniques, that is, a signal parameter estimation algorithm based on rotational invariance technology.

The theoretical basis of the present invention is as follows:

Assume that the antenna array of a DAR is a uniform linear array containing N antenna elements, the antenna element spacing is d, the carrier frequency of the transmitted signal is f ₀ , and the transmitted LFM signal is:

Where, s ₀ (t) is the transmitted linear frequency modulation signal; j is an imaginary unit, t is time, and T is pulse width; K=B/T is the frequency modulation slope of the LFM signal, B is the bandwidth of the LFM signal, and exp(·) is an exponential function.

Assume that there is a moving target at a certain position in the far field, with an initial distance of r ₀ , a target angle of θ, and a radial speed of v. Then, the mth echo signal x _n,m (t) received by the nth antenna element of the DAR can be expressed as:

Where n is the serial number of the DAR antenna element, n∈[0,N-1]; β is the backscatter coefficient of the point target; τn _,m is the round-trip delay. Assuming that within the radar coherent processing time range (a very short period of time), the target moves approximately in a uniform straight line in the radial direction, then,

Wherein, r _n (t _m ) represents the line of sight distance between the target and the nth antenna element when the mth pulse is emitted, which can be expressed as a function of the slow time t _m , where t _m = _mTr represents the time between pulses in the coherent processing process, m represents the sequence number of the pulse signal; _Tr represents the pulse repetition interval PRI of the radar emission signal; and c represents the speed of light.

The received signal xn _,m (t) is demodulated to obtain an intermediate frequency signal, which can be approximately expressed as:

Wherein, s _IF,n (t,t _m ) represents the intermediate frequency signal; * represents the conjugate of the complex number.

Then, the intermediate frequency signal is processed by pulse compression to obtain the pulse compressed signal, which can be expressed as:

In the formula, represents the pulse compression processing result of the mth pulse echo; α represents the complex amplitude of the signal after pulse compression, f _d = 2v/λ represents the Doppler frequency, and λ represents the wavelength of the transmitted signal. From formula (5), it can be seen that the pulse compression processing result of the echo signal of the moving target received by each antenna array element contains the angle θ of the target. If the target does not migrate within the coherent accumulation processing time range, that is, the radial movement distance of the target does not exceed the range resolution of the radar, the DOA (direction of arrival) of the target echo can be estimated by array signal processing technology.

Therefore, the signal direction of arrival estimation method based on constructing a received data matrix provided by the present invention specifically comprises the following steps:

Step 1: The number M of continuous pulse echoes that can be used for coherent accumulation processing by DAR must satisfy:

Where:

M—the number of consecutive pulse echoes that can be used for coherent accumulation processing;

c—speed of light;

B—Bandwidth of the LFM signal;

v— radial velocity of the target;

_Tr —The pulse repetition interval of the radar transmitting LFM signal.

Step 2: Demodulate the echo signal received by each antenna element of the DAR to obtain an intermediate frequency signal s _IF,n (t,t _m ):

Where:

s _IF,n (t,t _m )—intermediate frequency signal;

τ _n,m — round trip delay;

t—time;

T—pulse width;

β—backscatter coefficient of point target;

exp(·) — exponential function;

j—imaginary unit;

K—frequency modulation slope of LFM signal, K=B/T;

_f0 —transmitted signal carrier frequency;

r ₀ — initial distance of the target;

m—the sequence number of the pulse signal, m∈[1,M];

θ—target angle;

n—the serial number of the antenna array element of DAR, n∈[0,N-1], N is the number of antenna array elements;

d—antenna array element spacing.

Step 3: After digital sampling of the intermediate frequency signal s _IF,n (t,t _m ), a beamforming process is performed as shown in the following formula to obtain a beamforming output result, namely:

Where:

W—beamforming weight vector; used for spatial filtering to suppress interference and clutter in the spatial domain. A variety of conventional methods can be used to obtain the weight vector;

H—Conjugate transpose.

Step 4: Output the beamforming results Perform pulse compression processing to obtain the pulse compression processing result After processing M echo signals continuously, the processed data are arranged in parallel to obtain a data matrix

Step 5, perform Fourier transform processing on the data matrix obtained in step 4 in the slow time _tm domain to obtain the RD plane of the moving target.

Step 6, completing moving target detection on the RD plane obtained in step 5, thereby determining the distance position _ns where the moving target is located; wherein the moving target detection adopts a target detection algorithm, preferably a constant false alarm detection algorithm.

Step 7: demodulate and pulse compress the M echo signals received by the N array elements, collect the _nth complex values of each echo pulse compression result of each array element, and construct an N×M dimensional receiving data matrix X for DOA estimation, that is:

Step 8: For the constructed receiving data matrix X, it is known that it contains only one target echo signal, and a spatial spectrum estimation algorithm such as a multiple signal classification method or a rotation-invariant technology is used to obtain an estimated value of the arrival direction of the target echo.

In order to verify the feasibility and effectiveness of the method of the present invention, the following simulation test is carried out:

The number of antenna elements of DAR is N = 10, the carrier frequency of the transmitted LFM signal is f ₀ = 3 × 10 ⁹ Hz, the array element spacing d = 0.05m, the pulse width T = 10μs, the bandwidth of the radar transmitting LFM signal B = 6MB, the signal-to-noise ratio SNR = -35dB, the pulse repetition interval PRI = 200μs, the number of sampling points is 2048, the initial distance of the target r ₀ = 4500m, the radial motion speed v = 60m/s, the azimuth angle θ = 40°, the number of coherently accumulated sampling echoes M is 512, and the weight vector W of the DAR beamforming uses the array steering vector with a pointing angle θ′ = 45°, that is:

W=[1,exp(j2πdsinθ′/λ),…,exp(j2π(N-1)dsinθ′/λ)] ^T .

The simulation experiment results are shown in Figures 1 and 2, where Figure 1 is the RD plane and Figure 2 is the spatial spectrum estimation obtained by the MUSIC algorithm. It can be estimated that the target radial velocity is -60.05m/s and the target distance is 4504m. It can be seen from the simulation conditions that the detected target parameters are consistent with the actual ones.

Therefore, by performing target detection in the RD plane, the target distance position used to construct the data matrix can be obtained as n _s =1202, and the DOA of the target echo estimated by the constructed receiving data matrix is 40°, which is consistent with the actual azimuth angle θ of the target, thereby verifying the effectiveness of the method of the present invention. Compared with other methods, the method of the present invention can realize the estimation of the direction of arrival of non-stationary signals (LFM signals), and can accurately estimate the direction of arrival of the target echo under low signal-to-noise ratio (-35dB) conditions.

Claims (3)

1. A method for estimating the direction of arrival of a signal based on constructing a received data matrix, characterized in that it specifically comprises the following steps: Step 1: The number M of continuous pulse echoes that can be used for coherent accumulation processing by DAR must satisfy: Where: M—the number of consecutive pulse echoes that can be used for coherent accumulation processing; c—speed of light; B—Bandwidth of the LFM signal; v— radial velocity of the target; _Tr — the pulse repetition interval of the LFM signal transmitted by the radar; Step 2: Demodulate the echo signal received by each antenna element of the DAR to obtain an intermediate frequency signal s _IF,n (t,t _m ): Where: s _IF,n (t,t _m )—intermediate frequency signal; τ _n,m — round trip delay; t—time; T—pulse width; β—backscatter coefficient of point target; exp(·) — exponential function; j—imaginary unit; K—frequency modulation slope of LFM signal, K=B/T; _f0 —transmitted signal carrier frequency; r ₀ — initial distance of the target; m—the sequence number of the pulse signal, m∈[1,M]; θ—target angle; n—the serial number of the antenna array element of DAR, n∈[0,N-1], N is the number of antenna array elements; d—antenna array element spacing; Step 3: After digital sampling of the intermediate frequency signal s _IF,n (t,t _m ), a beamforming process is performed as shown in the following formula to obtain a beamforming output result, namely: Where: W—beamforming weight vector; H—conjugate transpose; Step 4: Output the beamforming results Perform pulse compression processing to obtain the pulse compression processing result After processing M echo signals continuously, the processed data are arranged in parallel to obtain a data matrix Step 5, performing Fourier transform processing on the data matrix obtained in step 4 in the slow time _tm domain to obtain the RD plane of the moving target; Step 6, completing the moving target detection on the RD plane obtained in step 5, thereby determining the distance position n _s where the moving target is located; Step 7: demodulate and pulse compress the M echo signals received by the N array elements, collect the _nth complex values of each echo pulse compression result of each array element, and construct an N×M dimensional receiving data matrix X for DOA estimation, that is: Step 8: For the constructed receiving data matrix X, it is known that it contains only one target echo signal, and a spatial spectrum estimation algorithm is used to process it to obtain an estimated value of the arrival direction of the target echo. 2. The signal direction of arrival estimation method based on constructing a received data matrix as described in claim 1, characterized in that in step 6, the moving target detection adopts a constant false alarm detection algorithm. 3. The signal direction of arrival estimation method based on constructing a received data matrix as described in claim 1, characterized in that in step 8, the spatial spectrum estimation algorithm is a multiple signal classification method or a rotation-invariant technology.