Radar target direction determination method, system and computer readable storage medium

Abstract

The present invention relates to a radar target direction determination method, system and computer-readable storage medium, and belongs to the field of radar technology. The radar target direction determination method of the invention is adopted, and the radar includes a transmitting antenna and three receiving antennas. The method uses the received echo signal to calculate the amplitude and phase of the reflected signals of the two targets after superposition, and uses a mathematical model to calculate the amplitude and phase of the reflected signals of the two targets before superposition, accurately estimates the wave direction of the two targets, and determines the target direction. Thereby effectively solving the problem of coherent signals, achieving high-resolution target positioning, and the calculation process in the radar target direction determination method of the present invention is simple, the corresponding hardware environment requirements are low, and the scope of application is quite wide.

Description

Technical Field

[0001] The present invention relates to the field of radar technology, and in particular to the field of radar target positioning technology, and specifically refers to a radar target direction determination method, system and computer-readable storage medium. Background Art

[0002] Radar can calculate the relative distance, relative speed, and direction of the target by emitting electromagnetic waves and receiving electromagnetic wave signals reflected by objects.

[0003] The direction of the target is estimated by measuring the direction of arrival of the electromagnetic wave reflected by the object. The existing practical direction of arrival estimation technologies are mainly divided into two categories. One is the algorithm represented by the linear prediction (LP) algorithm and the conventional beamforming (CBF) algorithm that directly processes the covariance matrix of the antenna array received data. The other is the algorithm represented by the multiple signal classification (MUSIC) algorithm and the rotationally invariant subspace (ESPRIT) algorithm that performs eigendecomposition on the covariance matrix of the antenna array received data.

[0004] When distinguishing signal sources from multiple directions, multiple receiving antennas must be designed. The angular resolution is proportional to the number of receiving antennas, and the number of distinguishable signal sources is proportional to the number of receiving antennas. In addition, in actual environments, the received data of signal sources from multiple directions generally have signal coherence problems, such as multipath phenomena during signal transmission. Coherent signal sources will make it impossible to correctly estimate the direction of arrival. At present, there are basically two categories of decoherence processing. One is a dimensionality reduction processing algorithm represented by spatial smoothing and matrix reconstruction. The other is a non-dimensionality reduction processing algorithm represented by frequency domain smoothing and Toeplite.

[0005] In order to meet the miniaturization requirements of vehicle-mounted radars, the antenna size, receiving antenna, and installation location are all limited by the installation environment. Therefore, the existing technology generally has problems such as low angular resolution and inability to resolve coherent signals. Summary of the invention

[0006] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a radar target direction determination method with high resolution, which can effectively solve the coherent signal problem, is simple to implement, has low requirements for the corresponding hardware environment, and has a wide range of applications.

[0007] In order to achieve the above-mentioned purpose, the radar target direction determination method of the present invention is applied to a variety of radar devices, especially suitable for millimeter wave radar, which includes an antenna array, and the antenna array includes a transmitting antenna and three receiving antennas. The radar target direction determination method includes the following steps:

[0008] (S101) The radar receives an echo signal;

[0009] (S102) performing two-dimensional fast Fourier transform processing on the echo signal to obtain a processing result;

[0010] (S103) extracting and obtaining a peak data model according to the processing results;

[0011] (S104) Calculate the direction of arrival based on the peak data model to determine the target direction.

[0012] In the radar target direction determination method, the three receiving antennas are not equidistant, the spacing between the first receiving antenna and the second receiving antenna is 0.5λ, and the spacing between the second receiving antenna and the third receiving antenna is 2.5λ, where λ is the wavelength of air.

[0013] In the radar target direction determination method, the step (S102) is specifically: performing a two-dimensional fast Fourier transform on the echo signal to convert the echo signal from time domain signal data to frequency domain signal data.

[0014] In the radar target direction determination method, the peak data model is specifically:

[0015] X=AS,

[0016] A=[a(θ1), a(θ2)],

[0017]

[0018] S = [ρ1, ρ2] T ,

[0019] Where X is the two-dimensional snapshot data vector; A is the two-dimensional flow matrix of the spatial array; S is the two-dimensional vector of the spatial signal; θ is the horizontal incident angle of the signal, [·] T is transposed.

[0020] In the radar target direction determination method, the step (S104) is specifically as follows:

[0021] Calculate Δ according to the following formula,

[0022] (α 2 -1)cos(14Δ)-(β 2 -1)cos(4Δ)+(β 2 -α 2 )cos(2Δ)=0

[0023] P is calculated according to the following formula:

[0024] (α 2 -1)P 2 -2[cos(4Δ)-α 2cos(2Δ)]P+α 2 -1=0

[0025] Calculate γ according to the following formula:

[0026]

[0027] Calculate ω1 and ω2 according to the following formula,

[0028] tanω m = -tan(mΔ)tanγ

[0029] Calculate Θ according to the following formula

[0030] η 21 =(ω2+2θ)-(ω1+θ)

[0031] η 21 is the phase difference;

[0032] Calculate φ1 and φ2 according to the following formula

[0033]

[0034]

[0035] Calculate the horizontal incident angle of the signal according to the following formula

[0036] φ1=-πsin(θ1)

[0037] φ2=-πsin(θ2)

[0038] The target direction is determined according to the horizontal incident angle of the signal.

[0039] The present invention also provides a computer-readable storage medium on which computer software is stored. When the computer software is read and executed, data interaction with the radar is realized, and the above-mentioned radar target direction determination method is realized.

[0040] The present invention also provides a radar target direction determination system, which is a vehicle-mounted system, and the radar is a vehicle-mounted millimeter wave radar. The system includes a CPU and a memory, and the memory stores computer software. When the CPU reads and runs the computer software, the system realizes data interaction with the radar and realizes the above-mentioned radar target direction determination method.

[0041] The radar target direction determination method of the invention is adopted. The radar includes a transmitting antenna and three receiving antennas. The method uses the received echo signal to calculate the amplitude and phase of the reflected signals of the two targets after superposition, and uses a mathematical model to calculate the amplitude and phase of the reflected signals of the two targets before superposition, accurately estimates the wave arrival direction of the two targets, and determines the target direction. Thereby effectively solving the problem of coherent signals and realizing high-resolution target positioning. In addition, the calculation process of the radar target direction determination method of the invention is simple, the corresponding hardware environment requirements are low, and the scope of application is quite wide. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The figure is a flow chart of the radar target direction determination method of the present invention.

[0043] Figure 2 The schematic diagram of the radar antenna layout in the radar target direction determination system of the present invention is shown in FIG. DETAILED DESCRIPTION

[0044] In order to more clearly understand the technical content of the present invention, the following embodiments are given in detail.

[0045] See also Figure 1 As shown, it is a schematic diagram of the flow chart of the radar target direction determination method of the present invention.

[0046] In one embodiment, the radar includes an antenna array, and the antenna array includes a transmitting antenna and three receiving antennas. Figure 2 As shown, the three receiving antennas are not equidistant, the spacing between the first receiving antenna and the second receiving antenna is 0.5λ, and the spacing between the second receiving antenna and the third receiving antenna is 2.5λ, where λ is the wavelength of air. The radar target direction determination method includes the following steps:

[0047] (S101) The radar receives an echo signal;

[0048] (S102) performing two-dimensional fast Fourier transform processing on the echo signal to obtain a processing result;

[0049] (S103) extracting and obtaining a peak data model according to the processing results;

[0050] (S104) Calculate the direction of arrival based on the peak data model to determine the target direction.

[0051] In a preferred embodiment, the step (S102) is specifically:

[0052] Performing a two-dimensional fast Fourier transform on the echo signal, converting the echo signal from time domain signal data to frequency domain signal data.

[0053] The peak data model is specifically:

[0054] X=AS,

[0055] A=[a(θ1), a(θ2)],

[0056]

[0057] S = [ρ1, ρ2] T ,

[0058] Where X is the two-dimensional snapshot data vector; A is the two-dimensional flow matrix of the spatial array; S is the two-dimensional vector of the spatial signal; θ is the horizontal incident angle of the signal, [·] T is transposed.

[0059] The step (S104) is specifically as follows:

[0060] Calculate Δ according to the following formula,

[0061] (α 2 -1)cos(14Δ)-(β 2 -1)cos(4Δ)+(β 2 -α 2 )cos(2Δ)=0

[0062] P is calculated according to the following formula:

[0063] (α 2 -1)P 2 -2[cos(4Δ)-α 2 cos(2Δ)]P+α 2 -1=0

[0064] Calculate γ according to the following formula:

[0065]

[0066] Calculate ω1 and ω2 according to the following formula,

[0067] tanω m = -tan(mΔ)tanγ

[0068] Calculate Θ according to the following formula

[0069] η 21 =(ω2+2θ)-(ω1+θ)

[0070] η 21 is the phase difference;

[0071] Calculate φ1 and φ2 according to the following formula

[0072]

[0073]

[0074] Calculate the horizontal incident angle of the signal according to the following formula

[0075] φ1=-πsin(θ1)

[0076] φ2=-πsin(θ2)

[0077] The target direction is determined according to the horizontal incident angle of the signal.

[0078] Correspondingly, the present invention also provides a computer-readable storage medium on which computer software is stored. When the computer software is read and executed, data interaction with the radar is realized, and the above-mentioned radar target direction determination method is realized.

[0079] In addition, the present invention also provides a radar target direction determination system, which can be a vehicle-mounted system, and the radar is a vehicle-mounted millimeter wave radar. The radar target direction determination system includes a CPU and a memory, and the memory stores computer software. When the CPU reads and runs the computer software, the system realizes data interaction with the radar and realizes the above-mentioned radar target direction determination method.

[0080] The present invention uses the signal data of the receiving antenna to obtain the amplitude and phase of the reflected signals from two targets in each channel after superposition. Then, using this known information, a new mathematical model is derived through the mathematical model of the received signal to calculate the amplitude and phase of the reflected signals of the two targets before superposition, so as to accurately estimate the wave arrival direction of the two targets.

[0081] In practical applications, Figure 2 Taking the millimeter-wave radar with one transmitting antenna and three receiving antennas as an example, the millimeter-wave radar uses one transmitting antenna to transmit electromagnetic waves, and uses three antennas to receive electromagnetic waves reflected back by surrounding targets, that is, the radar device receives the echo signal S101.

[0082] The two-dimensional FFT processing step S102 is to perform two-dimensional (distance dimension and speed dimension) fast Fourier transform processing on the received signal, and convert the received echo signal data from the time domain to the frequency domain, so as to estimate the relative distance and relative speed between the target and the radar.

[0083] The two-dimensional FFT peak extraction step S103 is to perform peak search based on the result of the two-dimensional FFT. The peak represents the target candidate, that is, there may be a target at the position of the distance dimension and the speed dimension where the peak is located. The peak standard data model (antenna array) after extraction is defined as shown in equations (1) to (5),

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] in is the L×1-dimensional snapshot data vector of the array. L is the number of receiving antennas. is the L×K dimensional flow matrix (steering vector array) of the spatial array. K is the number of targets at the same distance and speed relative to the radar, that is, the number of targets that need to be distinguished by the direction of arrival estimation algorithm. θ is the horizontal incident angle of the signal. λ is the air wavelength. d is the antenna spacing. Equation (3) is the uniform linear array model with equal spacing. is the K×1 dimensional vector (complex amplitude) of the spatial signal. is the L×1-dimensional noise vector of the spatial signal at time t. [·] T is transposed.

[0090] Based on the above peak standard data model, the resolution of two targets is taken as an example.

[0091] like Figure 2 As shown in the figure, the receiving antennas are unequally spaced. The spacing between Rx1 (receiving antenna No. 1) and Rx2 (receiving antenna No. 2) is 0.5λ, and the spacing between Rx2 and Rx3 (receiving antenna No. 3) is 2.5λ. The 0.5λ antenna spacing is set to solve the angle ambiguity problem. However, the 0.5λ antenna spacing is too close, and the angle estimation error is large in actual situations. Therefore, another set of 2.5λ antennas is designed.

[0092] This example only estimates the horizontal incident angle, assuming that the vertical angle and roll angle are 0. Based on the application scenario of vehicle-mounted millimeter-wave radar, radar data only calculates single-frame results without multi-frame accumulation processing.

[0093] According to the above settings, the standard model equations (1) to (5) are transformed into the following form:

[0094] X=AS (6)

[0095] A=[a(θ1), a(θ2)] (7)

[0096]

[0097] S = [ρ1, ρ2] T (9)

[0098] N = [n1, n2, n3] T (10)

[0099] Ignore the effect of noise N.

[0100] Substituting equations (7), (8), and (9) into equation (6), we obtain equation (11):

[0101]

[0102] For the sake of simplicity, let φ1 = -πsin(θ1) and φ2 = -πsin(θ2). Using Euler's formula to transform complex exponential into trigonometric function, we get formula (12)

[0103]

[0104] in Using the sine and cosine angle formula, we can further decompose formula (12) and get formula (13) after sorting it out:

[0105]

[0106] in Using the sine and cosine angle sum and difference formula, we can further decompose formula (13) and get formula (14) after sorting.

[0107]

[0108] Similarly, the other two items in the result matrix of formula (11) can be derived, and after sorting, formula (15) is obtained:

[0109]

[0110] Assume that equation (16) and equation (17) hold.

[0111] [cos(γ)cos(mΔ)-i sin(γ)sin(mΔ)]=M m e jω (16)

[0112] tanω m = -tan(mΔ)tanγ (17)

[0113] Then formula (15) can be transformed into formula (18)

[0114]

[0115] set up The above assumptions are taken as a system of equations, and the γ term is proposed after expansion, and equation (19) and equation (20) are obtained.

[0116]

[0117]

[0118] Combining equation (19) with equation (20) and eliminating the γ term, we get equation (21):

[0119] (α 2 -1)cos(14Δ)-(β 2 -1)cos(4Δ)+(β 2 -α 2 )cos(2Δ)=0 (21)

[0120] Here, only Δ is an unknown in equation (21), so an approximate solution for Δ can be obtained by solving the equation. Then, the calculated Δ is used to further calculate the ratio of the amplitude of the received signal. Known Using the sine-cosine conversion formula and the sine-double angle formula, cos(2γ) can be rearranged to obtain formula (22):

[0121]

[0122] Combining equation (22) with equation (19), we can get equation (23)

[0123] (α 2 -1)P 2 -2[cos(4Δ)-α 2 cos(2Δ)]P+α 2 -1=0 (23)

[0124] Here, only P is an unknown number in equation (23), so the solution of P can be obtained by solving the equation. At this time, γ can be calculated by equation (22) using the value of P, and the values ​​of ω1 and ω2 can be further calculated using equation (17). Then, the phase difference η can be obtained by the known number 21 The value of Θ found by solving equation (24).

[0125] η 21 =(ω2+2Θ)-(ω1+Θ) (24)

[0126] Finally, using the calculated Δ and Θ, φ1 and φ2 can be derived, thus obtaining the final results θ1 and θ2.

[0127] Compared with the prior art, the radar target direction determination method of the present invention has the following advantages:

[0128] 1. High angular resolution. The test result of angular resolution is within 1 degree. The test is divided into two parts: simulation test and prototype test. In the simulation test, the normal direction of the radar is set to 0 degrees, and any two targets with an angle difference of 1 degree are selected in the range of -75 degrees to 75 degrees for testing. The final test results all achieve a resolution of less than 1 degree. In the prototype test, two targets (corner reflectors) are set in front of the radar. The two stationary targets are at the same distance and the angles are gradually separated until the radar can distinguish them. The final test results all achieve a resolution of less than 1 degree.

[0129] 2. Effectively solve the coherent signal problem. Under the influence of coherent signals, the azimuth angle of the target can still be calculated.

[0130] 3. The antenna layout is unequal spacing. It is proved that the method of the present invention has no special restrictions on the antenna spacing, and it can be equal spacing or unequal spacing. Therefore, it can be widely used in various radar devices to meet the demand for radar miniaturization.

[0131] 4. The calculation process is simple and requires low hardware resources. The actual test prototype uses the low-cost AWR1642 chip (CPU: ARM-Cortex R4F 200MHz, DSP: C674x DSP 600MHz, on-chip memory: 1.5MB). Since the calculation process of the present invention is mostly equation calculation, there is no cyclic calculation processing with large amount of calculation. Therefore, the operation of the calculation process can be fully realized using limited computing resources and storage resources.

[0132] The radar target direction determination method of the invention is adopted. The radar includes a transmitting antenna and three receiving antennas. The method uses the received echo signal to calculate the amplitude and phase of the reflected signals of the two targets after superposition, and uses a mathematical model to calculate the amplitude and phase of the reflected signals of the two targets before superposition, accurately estimates the wave arrival direction of the two targets, and determines the target direction. Thereby effectively solving the problem of coherent signals and realizing high-resolution target positioning. In addition, the calculation process of the radar target direction determination method of the invention is simple, the corresponding hardware environment requirements are low, and the scope of application is quite wide.

[0133] In this specification, the present invention has been described with reference to specific embodiments thereof. However, it is apparent that various modifications and variations may be made without departing from the spirit and scope of the present invention. Therefore, the specification and drawings should be regarded as illustrative rather than restrictive.

Claims

1. A method for determining a radar target direction, characterized in that: The radar includes an antenna array, the antenna array includes a transmitting antenna and three receiving antennas, the three receiving antennas are not equidistant, the spacing between the first receiving antenna and the second receiving antenna is 0.5λ, the spacing between the second receiving antenna and the third receiving antenna is 2.5λ, λ is the wavelength of air, and the radar target direction determination method includes the following steps: (S101) The radar receives an echo signal; (S102) performing a two-dimensional fast Fourier transform process on the echo signal to convert the echo signal from time domain signal data into frequency domain signal data; (S103) extracting and obtaining a peak data model according to the processing result, wherein the peak data model is specifically: X=AS, A=[a(θ1),a(θ2)], S=[ρ1,ρ2] T , Where X is the two-dimensional snapshot data vector; A is the two-dimensional flow matrix of the spatial array; S is the two-dimensional vector of the spatial signal; θ is the horizontal incident angle of the signal, [·] T is transposed; (S104) Calculating the direction of arrival based on the peak data model to determine the target direction is: Calculate Δ according to the following formula: (a 2 ―1)cos(14Δ)―(β 2 ―1)cos(4Δ)+(β 2 -a 2 )cos(2Δ)=0 Calculate P according to the following formula, where P is the ratio of the amplitude of the received signal (a 2 ―1)P 2 ―2[cos(4Δ)―α 2 cos(2Δ)]P+α 2 ―1=0 Calculate γ according to the following formula: Calculate ω1 and ω2 according to the following formula, I'm going to m =―tan(,Δ)tanγ Calculate Θ according to the following formula or 21 =(ω2+2Θ)―(ω1+Θ) η 21 is the phase difference; Calculate φ1 and φ2 according to the following formula Calculate the horizontal incident angle of the signal according to the following formula φ1=―πsin(θ1) π2=―πsin(θ2) The target direction is determined according to the horizontal incident angle of the signal.

2. A computer-readable storage medium, characterized in that: Computer software is stored thereon, and when the computer software is read and run, data interaction with the radar is realized, and the radar target direction determination method described in claim 1 is realized.

3. A radar target direction determination system, characterized in that: The system comprises a CPU and a memory, wherein the memory stores computer software. When the CPU reads and runs the computer software, the system realizes data interaction with the radar and realizes the radar target direction determination method according to claim 1.

4. The radar target direction determination system as claimed in claim 3, characterized in that: The system is a vehicle-mounted system, and the radar is a vehicle-mounted millimeter-wave radar.

Images