CN-118655564-B - Method for repositioning moving targets in building based on unmanned aerial vehicle-mounted double-channel through-wall radar interference phase

CN118655564BCN 118655564 BCN118655564 BCN 118655564BCN-118655564-B

Abstract

The invention discloses a method for repositioning a moving target in a building based on an unmanned aerial vehicle-mounted double-channel through-wall radar interference phase, which comprises the step of providing a novel method for repositioning the azimuth of the moving target based on the double-channel interference phase. The method firstly extracts the two-channel interference phase of the moving target in a distance compression domain. And secondly, establishing an interference phase polynomial model, and calculating coefficients of each order by using a least square method. Finally, the obtained polynomial coefficients are used for constructing and solving an optimization problem about the azimuth position of the moving target. In contrast to conventional methods, the present invention does not require imaging and multi-channel registration operations. In addition, the method provided by the invention can directly obtain the azimuth position of the moving target without estimating the radial speed and calculating the offset, thereby avoiding the problem of space Doppler ambiguity of the radial speed estimation faced by the traditional synthetic aperture radar moving target positioning technology. The validity of the proposed method is verified through simulation.

Inventors

QU XIAODONG
YANG XIAOPENG
ZHANG HAO
LIU FEIYANG
SUN XIAOLONG
MA ZEYU

Assignees

北京理工大学

Dates

Publication Date: 20260512
Application Date: 20240612

Claims (11)

1. A method for repositioning a moving target of an unmanned aerial vehicle through-wall radar based on a double-channel interference phase is characterized by comprising the following steps: Step S1, acquiring original echo data of a human body target of an unmanned aerial vehicle through-wall radar, and performing pulse compression along a distance direction; S2, deriving a function relation between the discrete interference phase and slow time, constructing a polynomial regression model, and estimating each order coefficient by using a least square method; s3, constructing an optimization function related to azimuth and position, and solving by using pattern searching; The interference phase function is (1); Wherein, the The representative wavelength of the light is represented by, Corresponding to slow time The instantaneous tilt of the transmitting antenna to the moving object, Representing the radial velocity of the moving object: in the formula (1), constant terms of the interference phase function comprise azimuth position information of a moving target and distance information of extra propagation of electromagnetic waves in a wall body, and components caused by the wall body in the interference phase are as follows: (2); Wherein, the ; Wherein, the Representing the thickness of the wall; In the through-wall scene, the method comprises the steps of, On the order of 10-4, whereas the two-channel skew difference in the constant term in equation (2) is part of On the order of 10-2, hence here Neglecting; interferometric phase expression rewrites as (3); Wherein, the Representing the double-pass distance of the electromagnetic wave from the transmitting antenna to the moving target and then to the mth receiving antenna; in the formula (3), after the conjugate multiplication of adjacent channels, the influence of the wall body is eliminated, and the interference phase constant term is only matched with , And Related to; Image indexing slow time by distance compressed domain Corresponding to the single snapshot data acquisition; the invention uses polynomial regression model to fit because the extracted interference phase is a quadratic function of slow time, the formula (3) is a nonlinear interference phase model, and the discrete form is written as (4); Where j is a discrete time variable, Representing the number of azimuth sampling points, Representing noise in matrix form expressed as ; Wherein, the ; ; ; ; Wherein the method comprises the steps of Representing a matrix transposition operation; representing polynomial coefficients of each order, and obtaining by using least square method joint estimation: 。
2. The method of claim 1, wherein the radar emits a chirped continuous wave (Linear Frequency Modulated, LFM) signal, i.e (5); Wherein, the Representing a fast time variable which is a function of the time, As a result of the center frequency, In order to achieve a frequency modulation rate, In order to be a pulse width, As window function, scattering the moving point target after passing through the wall, the first The echoes received by the receiving antennas are as follows: (6); wherein c represents the speed of light, Is the scattering coefficient of the moving object.
3. The method of claim 1, wherein the received echo signal is demodulated to baseband in the form of: (7); Wherein, the Is the scattering coefficient of the moving object, Is the speed of light.
4. The method according to claim 1, wherein after distance pulse compression, a pulse is obtained (8); Wherein, the In order to transmit the bandwidth of the signal, Representing wavelength; representing the double-pass distance of electromagnetic wave propagation (9); Wherein, the Is the first Instantaneous distance of the individual receiving antennas to the moving object: (10); For the instantaneous distance of the transmitting antenna to the moving object: (11); Wherein, the Respectively, the transmitting antenna is at slow time Azimuth position and distance position at time, and , Corresponding to slow time Instantaneous slant distance from the moment transmitting antenna to the moving target; representing the extra propagation path of the electromagnetic wave inside the wall.
5. The method of claim 1, wherein the incident direction of the emitted electromagnetic wave forms an angle with the plane of the wall (12); The included angle between the directions 1 and 2 of the electromagnetic wave from the target return receiving antenna and the wall body is that (13); Then (14); Wherein, the Representing the thickness of the wall body, Representing the dielectric constant of the wall, in particular, the free space scene corresponds to 。
6. The method of claim 1, wherein distance expressions (10) and (11) are defined in And (5) performing Taylor expansion at the moment to obtain: (15); (16); the expansion of (9) is written as (17); Wherein, the 。
7. The method of claim 1, wherein the doppler frequency of the moving object is obtained from equation (17) (18); Wherein, the Is the doppler frequency shift caused by the radial velocity of the moving target: (19); is the radial velocity of the moving object: ; Is the Doppler frequency caused by platform motion and receive antenna azimuth position (20); Doppler frequency shift due to wall refraction: (21); the azimuth frequency modulation rate of the moving target is: (22); azimuth frequency modulation rate for radar platform motion induced: (23); In particular, for stationary objects, , After azimuth compression, the Doppler frequency offset of the moving target is accurately focused to a real azimuth position, compared with a static target, the Doppler frequency offset of the moving target is offset, and according to the (21), the influence of the wall body on the Doppler frequency offset of the moving target is 0, and only the offset caused by radial speed is required to be compensated According to formula (19), Radial velocity with moving object And azimuth position All related.
8. The method of claim 1, wherein when satisfied In the case of the conditions of (2), The azimuth offset is (24); Thereby obtaining the true azimuth position (25); In a through-wall scene, the method can not meet Conditions that result in systematic errors in the conventional relocation method.
9. The method of claim 1, wherein the results of the two receive channels are conjugate multiplied in a distance compression domain to obtain (26)。
10. The method of claim 1, wherein to obtain the moving target azimuth position, the following optimization problem is established: (27); Wherein, the (28)。
11. The method of claim 1, wherein equation (27) is solved using a pattern search algorithm to ultimately estimate the moving target azimuth position.

Description

Method for repositioning moving targets in building based on unmanned aerial vehicle-mounted double-channel through-wall radar interference phase Technical Field The invention relates to the field of radar signal processing, in particular to a method for repositioning a moving target of an unmanned airborne through-wall radar. Background In the scenes of urban street fighting, post-disaster rescue and the like, the accurate acquisition of the target position after the obstacle is beneficial to making tactical planning, mastering the combat advantages and improving the rescue efficiency. The unknown radial speed of the moving target can cause the imaging position to shift, and the real position cannot be obtained directly through imaging. The invention provides a repositioning technology of a moving target in a building, which aims to provide a technology capable of accurately estimating the azimuth position of the moving target in a shielding space. Conventional multi-channel moving target azimuth repositioning methods include a speed synthetic aperture radar (Velocity Synthetic Aperture Radar, VSAR), a track-along-the-track interferometry (Along-track Interferometry, ATI), an adaptive matched filtering (ADAPTIVE MATCHED FILTER, AMF), and a Subspace Projection (SP). The methods all utilize the multichannel moving target phase to estimate the radial speed of the moving target, calculate the azimuth position offset and realize repositioning. However, conventional moving object radial velocity estimation methods require strict registration of the multi-channel images. By introducing a joint pixel signal model and utilizing multi-channel auxiliary pixel information and combining a traditional radial velocity estimation method, the motion parameter accurate estimation of channel registration errors can be realized. However, the method has higher spatial freedom degree with higher requirements depending on the number of channels, and the estimation accuracy is limited under the double channels. Based on the above research, we consider that the existing method mostly utilizes the phase information of the multi-channel moving target to estimate the radial velocity, and calculates the azimuth position offset to finish repositioning. This requires that the platform speed be much greater than the speed of the moving object and the shortest tilt be much greater than the moving object azimuth position coordinates, and some methods also require that the channel registration conditions of the offset phase center antenna (DISPLACED PHASE CENTER ANTENNA, DPCA) be met. In the through-wall detection scene, the azimuth position, the pitch course, the platform speed and the moving target speed are all in the same order of magnitude, so that the precision of the traditional moving target repositioning method is reduced. In order to solve the above problems, we propose a new moving target azimuth repositioning method. The method utilizes interference phase function extraction and fitting to establish an optimization problem, adopts the mode search optimizing technology and the like to obtain the azimuth position of the moving target, and can improve the positioning accuracy of the moving target. Disclosure of Invention The invention provides a moving target repositioning method based on an unmanned aerial vehicle through-wall radar double-channel interference phase, which comprises the following steps: Step S1, acquiring original echo data of a human body target of an unmanned aerial vehicle through-wall radar, and performing pulse compression along a distance direction; S2, deriving a function relation between the discrete interference phase and slow time, constructing a polynomial regression model, and estimating each order coefficient by using a least square method; and S3, constructing an optimization function about azimuth and position, and solving by using pattern search. The radar emits a chirped continuous wave (Linear Frequency Modulated, LFM) signal, i.e Where τ represents the fast time variable, f 0 is the center frequency, K r is the frequency modulation, T p is the pulse width, and rect (·) is the window function. The echoes received by the mth receiving antenna after scattering by the moving point target behind the wall are as follows: the form of the received echo signal after demodulation to baseband is: Wherein σ t is the scattering coefficient of the moving object, and c is the speed of light. After distance pulse compression, can obtain Where b=k rTp is the transmit signal bandwidth and λ represents the wavelength. R 0m(ta) represents the double-pass distance of electromagnetic wave propagation R0m(ta)＝R(ta)+Rm(ta)+Rwall,0m (5) Wherein, R m(ta) is the instantaneous distance from the mth receiving antenna to the moving object: r (t a) is the instantaneous distance of the transmitting antenna to the moving object: Where x (t a),y(ta) is the horizontal position of the transmitting antenna at the time of slow time