Terahertz SAR two-dimensional self-focusing imaging algorithm

Abstract

The THz SAR two-dimensional self-focusing imaging algorithm comprises the following steps in sequence: a THz SAR echo signal model is established; a residual video phase term and an inclined phase term are removed; a distance direction nonlinear phase error is estimated; distance direction compensation is performed based on the estimated nonlinear phase error; a distance compressed echo signal is obtained after distance direction compensation; a distance direction coarse compensation is performed on the distance compressed echo signal; a range migration correction is performed on the distance direction coarse compensation echo signal; a residual phase error in the azimuth direction is estimated; azimuth direction fine compensation is performed based on the estimated residual phase error; azimuth direction fine compensation is performed on the azimuth direction fine compensation echo signal; and a THz-SAR focusing image is obtained after azimuth direction compression. In the present application, the distance direction nonlinear phase error and the azimuth direction motion phase error are estimated and compensated to obtain a high-resolution THz-SAR focusing image; and compared with directly using the self-focusing algorithm, the joint use of IMU / GPS measurement data and the MEA self-focusing algorithm can reduce the amount of calculation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar signal processing, and particularly relates to a two-dimensional self-focusing imaging algorithm for terahertz SAR. BACKGROUND

[0002] Terahertz (THz) waves refer to electromagnetic waves with a frequency spectrum between 100 GHz and 10 THz, and have the characteristics of high carrier frequency, large communication capacity, good penetration, low photon energy, and no biological ionization. Compared with microwave synthetic aperture radar (SAR) imaging, terahertz synthetic aperture radar (THz-SAR) imaging has the advantages of higher resolution, higher frame rate, higher detection probability, and easier identification, so it has attracted more and more attention in the field of modern radar imaging.

[0003] When THz-SAR imaging is performed, it is usually affected by some non-ideal factors, which introduce phase errors in the radar echo signal. The error can be divided into two parts in the range direction and the azimuth direction. The range direction phase error is caused by the fact that the solid-state source based terahertz radar system can only generate terahertz signals by multiple frequency multiplication, and the non-ideal characteristics of the frequency sweeping source and the transceiver link frequency multiplier introduce nonlinear phase errors in the terahertz signal. These nonlinear phase errors will directly affect the phase of the target echo, resulting in distortion of the range image and degradation of the resolution, causing the target support area to not conform to the real situation. The azimuth direction phase error is caused by the fact that the flight trajectory deviates from the ideal state due to the influence of air flow and other factors on the platform. In modern SAR systems, the motion state data of the platform can be measured by combining the inertial measurement unit (IMU) and the global positioning system (GPS), and then the echo is compensated by the measurement data. However, for THz-SAR, the short wavelength makes the influence of small vibration errors on the echo phase cannot be ignored, which makes the existing sensors cannot meet the accuracy requirements of compensation. Therefore, for the above-mentioned range direction and azimuth direction phase errors, after compensation based on measurement data, a self-focusing algorithm is still needed to improve the imaging quality.

[0004] Map Drift (MD) and Phase Gradient Autofocus (PGA) are two important autofocusing algorithms. The MD method has the disadvantage of only estimating quadratic phase error, and the improved multi-subaperture correlation algorithm can estimate high-order phase error in principle, but the estimation result is more biased due to the too short subaperture, and often cannot meet the requirements of high-resolution imaging. PGA has good robustness and high imaging efficiency, but must have a special highlight to effectively compensate for errors, which has limitations in practical applications. The Minimum Entropy Algorithm (MEA) algorithm is based on the overall image entropy value without special highlights, and has a wider application range and more practicality. SUMMARY

[0005] The purpose of the present application is to provide a terahertz SAR two-dimensional autofocusing imaging algorithm, which obtains a THz-SAR high-resolution focused image by the combination of the measurement data-based compensation and the MEA-based two-dimensional autofocusing algorithm.

[0006] In order to achieve the above purpose, the present application provides a terahertz SAR two-dimensional autofocusing imaging algorithm, comprising the following steps:

[0007] S1: establishing a model of echo signals of terahertz SAR;

[0008] S2: removing residual video phase terms and oblique phase terms in the echo signals;

[0009] S3: estimating nonlinear phase error in the range direction;

[0010] S4: performing range compensation based on the estimated nonlinear phase error;

[0011] S5: performing range compression on the echo signals after range compensation;

[0012] S6: performing azimuth coarse compensation on the echo signals after range compression based on IMU / GPS measurement data;

[0013] S7: performing range migration correction on the echo signals after azimuth coarse compensation;

[0014] S8: estimating residual phase error in the azimuth direction;

[0015] S9: performing azimuth fine compensation based on the estimated residual phase error;

[0016] S10: performing azimuth compression on the echo signals after azimuth fine compensation to obtain a THz-SAR focused image.

[0017] Further, in the two-dimensional self-focusing imaging algorithm of the terahertz SAR, in step S1, the terahertz SAR transmits a linear frequency modulation pulse signal, and performs linear frequency modulation on a received echo signal, and a model expression of the echo signal generated is:

[0018]

[0019] Wherein, τ is a distance direction fast time; t is an azimuth direction slow time; T p is a pulse width; c is a light speed; λ is a wavelength; R i is an actual distance of the radar; R ref is a reference distance of the radar; R Δ is a difference between the actual distance and the reference distance of the radar; j is an imaginary unit; γ is a frequency modulation slope.

[0020] Further, in the two-dimensional self-focusing imaging algorithm of the terahertz SAR, in step S2, a Fourier transform (FFT) is performed on the distance direction of formula (1), and the following formula (2) is obtained:

[0021]

[0022] Wherein, f r is a distance direction frequency; f c is a center frequency; is a Doppler term; is a residual video phase term; is a squint phase term of an echo envelope;

[0023] The residual video phase term and the squint phase term in formula (2) are expressed as:

[0024]

[0025] Formula (2) is multiplied by a compensation function of the residual video phase term and the squint phase term The following formula (4) is obtained:

[0026]

[0027] An inverse Fourier transform (IFFT) is performed on the fast time of formula (4), and a model expression of the echo signal after the transform is:

[0028]

[0029] Further, in the two-dimensional self-focusing imaging algorithm of the terahertz SAR, in step S3, it is assumed that a function of the nonlinear phase error in the distance direction is Then, a model expression of the echo signal with the nonlinear phase error is:

[0030]

[0031] where τ is the range direction fast time; t is the azimuth direction slow time;

[0032] Discretize τ and t as t n (n = 0: N - 1) and t m (m = 0: M - 1), then the discretization of equation (6) can be obtained as:

[0033]

[0034] The minimum entropy criterion is taken as the evaluation criterion, and the estimation of the nonlinear phase error is carried out through iterative optimization, and the specific steps are as follows:

[0035] S3.1: assuming that the estimated nonlinear phase error of the range direction is and The initialization is set to 0, that is, the expression of the compensated range image is:

[0036]

[0037] where k is the range frequency; m is the number of echo pulses; n is the range unit; N is the total number of range units;

[0038] S3.2: based on the minimum entropy criterion, a model of the phase error is established, and the expression is:

[0039]

[0040] where Ent is the image entropy value, and the expression is:

[0041]

[0042] where E is the range image energy, and the expression is:

[0043]

[0044] S3.3: based on the Newton method to solve equation (9), the expression of the iterative phase error can be obtained as:

[0045]

[0046] where:

[0047]

[0048]

[0049] where s0 (l-1)(n, m) is the echo data corrected by the phase error estimated by the (l-1)th iteration; G is the complex conjugate of G k，m each element in G is the complex conjugate of the corresponding element in G

[0050] S3.4: Determine whether the estimated phase error is accurate enough, i.e., whether the difference between the image entropy values Ent and Ent (l-1) obtained after the lth and (l-1)th iterations, respectively, is lower than a pre-set threshold value J, i.e.:

[0051] |Ent (l) -Ent (l-1) |≤J (13)

[0052] If formula (13) is satisfied, the nonlinear phase error of the range direction that needs to be compensated is:

[0053]

[0054] In step S4, the nonlinear phase error of the range direction is compensated according to the nonlinear phase error estimated in step S3 The compensation function of the nonlinear phase error of the range direction is: and multiplied by formula (7) to compensate the echo signal in the range direction, and the following formula can be obtained:

[0055]

[0056] Further, in the two-dimensional self-focusing imaging algorithm of the terahertz SAR, in step S5, the echo signal after range direction compensation is subjected to a range direction Fourier transform (FFT), and the model expression of the echo signal after range compression is:

[0057]

[0058] where t i is the discrete delay fast time of the target to the radar; is the discrete slow time of the beam center passing through the target; B is the range direction signal bandwidth; w a (·) is the azimuth direction envelope;

[0059] Based on the existence of the motion phase error , the model expression of the echo signal after range compression is:

[0060]

[0061] Further, in the two-dimensional self-focusing imaging algorithm of the terahertz SAR, in step S6, based on the IMU / GPS measurement data, the difference d between the actual slant range of the antenna and the scene and the ideal slant range is calculated and obtainedlos (t n , t m ) to make corresponding motion phase compensation, and the motion phase error compensation function is:

[0062]

[0063] Formula (17) is multiplied by formula (18) to make azimuth direction rough compensation to the echo signal, and the following can be obtained:

[0064] s a，com1 (t n , t m ) = s a (t n , t m) · D com

[0065]

[0066] Based on the residual phase error in the azimuth direction due to the IMU / GPS measurement data Transforming formula (19) can obtain:

[0067]

[0068] Further, in the two-dimensional self-focusing imaging algorithm of the terahertz SAR, in step S7, the echo signal after azimuth direction rough compensation is subjected to azimuth direction Fourier transform (FFT), the echo signal is transformed to the range Doppler domain, the echo signal is subjected to range migration correction through interpolation, and after the influence of the range migration is eliminated, the echo signal is subjected to inverse azimuth direction Fourier transform (IFFT), and the following can be obtained:

[0069]

[0070] Wherein: is the discrete delay fast time of the target to the radar nearest slant range.

[0071] Further, in the two-dimensional self-focusing imaging algorithm of the terahertz SAR, in step S8, based on the minimum entropy criterion, the residual phase error in the azimuth direction is estimated through the phase error model in step S3.2, after l times of iteration, it is judged whether the image entropy value Ent obtained satisfies formula (13), if yes, the residual phase error in the azimuth direction that needs to be compensated is:

[0072]

[0073] In step S9, according to the residual phase error estimated in step S8 The compensation function of the residual phase error in the azimuth direction is and multiplied by formula (21) to perform azimuth fine compensation on the echo signal, and the following formula (22) is obtained:

[0074]

[0075] Further, in the THz SAR two-dimensional self-focusing imaging algorithm, the matched filtering is performed on the fine-compensated echo signal in step S10, and after the azimuth compression, the model expression of the THz-SAR focusing image obtained is as follows:

[0076]

[0077] Wherein, B d is the Doppler bandwidth.

[0078] Compared with the prior art, the beneficial effects of the present application mainly lie in: the non-linear phase error in the range direction and the motion phase error in the azimuth direction are comprehensively considered, and corresponding compensation is performed after estimation, so that a high-resolution THz-SAR focusing image is obtained; at the same time, based on the joint use of IMU / GPS measurement data and MEA self-focusing algorithm, compared with directly using the self-focusing algorithm, the operation amount is effectively reduced. BRIEF DESCRIPTION OF DRAWINGS

[0079] Figure 1 It is a structure schematic diagram of the THz SAR two-dimensional self-focusing imaging algorithm in the present application. DETAILED DESCRIPTION

[0080] The THz SAR two-dimensional self-focusing imaging algorithm of the present application will be described in more detail below in conjunction with the schematic diagram, wherein the preferred embodiments of the present application are represented, and it should be understood that the present application described herein can be modified by those skilled in the art, and the advantageous effects of the present application can still be achieved. Therefore, the following description should be understood as extensive knowledge for those skilled in the art, and not as a limitation on the present application.

[0081] The present application will be described in more detail in the following paragraphs with reference to the accompanying drawings. The advantages and features of the present application will be more apparent according to the following description. It should be noted that the drawings are all greatly simplified and use non-precise proportions, only for the purpose of facilitating and clearly assisting the description of the embodiments of the present application.

[0082] As Figure 1 shown, the present application proposes a THz SAR two-dimensional self-focusing imaging algorithm, including the following steps:

[0083] S1: establishing a model of the echo signal of the THz SAR;

[0084] In step S1, the terahertz SAR transmits a linear frequency modulation pulse signal, and linear frequency modulation is performed on the received echo signal, and the model expression of the generated echo signal is:

[0085]

[0086] Wherein: τ is the distance fast time; t is the azimuth slow time; T p is the pulse width; c is the speed of light; λ is the wavelength; R i is the actual distance of the radar; R ref is the reference distance of the radar; RΔ is the difference between the actual distance and the reference distance of the radar; j is the imaginary unit; γ is the frequency modulation slope;

[0087] S2: remove the residual video phase term and the inclined phase term in the echo signal;

[0088] In step S2, the Fourier transform of formula (1) in the distance direction is performed, and the following formula (2) is obtained:

[0089]

[0090]

[0091] Wherein: f r is the distance frequency; f c is the center frequency; is the Doppler term; is the residual video phase term; is the inclined phase term of the echo envelope; Since the residual video phase term and the inclined phase term will cause the Doppler value to deviate, it is necessary to remove the two terms;

[0092] The residual video phase term and the inclined phase term in formula (2) are expressed as:

[0093]

[0094] The compensation function of the residual video phase term and the inclined phase term is And multiply formula (2) to obtain:

[0095]

[0096] The inverse Fourier transform (IFFT) of formula (4) in the fast time is performed, and the model expression of the transformed echo signal is:

[0097]

[0098] S3: estimate the nonlinear phase error in the distance direction based on MEA;

[0099] In step S3, the linear phase is not ideal due to the existence of the nonlinear phase error in the range direction, which will affect the distribution and focusing of the range image. Assuming that the function of the nonlinear phase error in the range direction is The model expression of the echo signal with the nonlinear phase error is

[0100]

[0101] Discretize the fast time τ in the range direction and the slow time t in the azimuth direction as t n (n = 0: N - 1) and t m (m = 0: M - 1), respectively, and discretize equation (6) to obtain

[0102]

[0103] Take the minimum entropy as the evaluation criterion, and estimate the nonlinear phase error by iterative optimization. The specific steps are as follows:

[0104] S3.1: Assume that the estimated nonlinear phase error in the range direction is and Initialize to 0, that is, the expression of the compensated range image is

[0105]

[0106] where k is the distance frequency, m is the number of echo pulses, n is the distance unit, and N is the total number of distance units;

[0107] S3.2: Based on the minimum entropy criterion, establish the optimization model of the phase error, and the expression is

[0108]

[0109] where Ent is the image entropy value, and the expression is

[0110]

[0111] where E is the range image energy, and the expression is

[0112]

[0113] S3.3: Based on the Newton method, solve equation (9) to obtain the expression of the iterative phase error

[0114]

[0115] where

[0116]

[0117]

[0118] wherein: s0 (l-1 ) is the echo data corrected by the phase error estimated in the (l-1)th iteration: is G k，m each element is complex conjugate;

[0119] S3.4: judging whether the estimated phase error is accurate enough, i.e. judging whether the difference between the image entropy values Ent and Ent (l-1 ) respectively calculated after the lth and (l-1)th iterations is lower than a pre-set threshold value J, i.e.:

[0120] |Ent (l) -Ent (l-1) |≤J (13)

[0121] wherein: the smaller the pre-set threshold value J is, the more accurate the estimated phase error is, but the iteration number will increase accordingly;

[0122] if the formula (13) is satisfied, the nonlinear phase error of the range direction which needs to be compensated is:

[0123]

[0124] S4: compensating the range direction based on the estimated nonlinear phase error;

[0125] In step S4, according to the nonlinear phase error of the range direction estimated in step S3, the compensation function of the nonlinear phase error of the range direction is: and multiplied by the formula (7) to compensate the echo signal in the range direction, and the model expression of the echo signal is:

[0126]

[0127] S5: compressing the range direction of the echo signal compensated in the range direction;

[0128] In step S5, the echo signal compensated in the range direction is subjected to Fourier transform (FFT) in the range direction to complete the range compression, i.e. the model expression of the echo signal is:

[0129]

[0130] wherein: t i is the discrete delay fast time of the target to the radar; is the discrete slow time of the beam center passing through the target; B is the bandwidth of the range direction signal; w a (·) is the azimuth direction envelope; R Δthe difference between the actual distance of the radar and the reference distance, i.e. the ideal slant range;

[0131] Based on the existence of the motion phase error , the model expression of the echo signal actually completing the range compression is:

[0132]

[0133] S6: Based on the IMU / GPS measurement data, the range-compressed echo signal is azimuthally coarsely compensated;

[0134] In step S6, based on the IMU / GPS measurement data, the actual trajectory of the antenna is calculated in combination with the geometric position of the antenna, and the reference trajectory of the antenna is obtained through linear fitting; and the actual slant range and the ideal slant range of the antenna with the scene are respectively calculated according to the actual trajectory and the reference trajectory of the antenna, so as to obtain the difference d los (t n , t m ) between the actual slant range and the ideal slant range, and the corresponding motion phase compensation is performed for the motion phase error , and the compensation function of the motion phase error is:

[0135]

[0136] Formula (17) is multiplied by formula (18) to coarsely compensate the echo signal in the azimuth direction, and the following can be obtained:

[0137]

[0138] In formula (19), when is 0, the phase error in the azimuth direction is completely compensated; however, due to the limitation of the IMU / GPS measurement data, residual phase error in the azimuth direction will exist. Therefore, formula (19) is transformed, and the following can be obtained:

[0139]

[0140] S7: The echo signal after the azimuthal coarse compensation is corrected for range migration;

[0141] In step S7, the echo signal after the azimuthal coarse compensation is subjected to azimuthal Fourier transform (FFT), at this time the echo signal is transformed into the range-Doppler domain, the echo signal is corrected for range migration through interpolation, and after the influence of the range migration is eliminated, the echo signal is subjected to inverse azimuthal Fourier transform (IFFT), and the following can be obtained:

[0142]

[0143] wherein: The discrete delay fast time of the target to the radar's nearest slant range; it is known that after the range migration correction, the azimuth is still affected by the residual phase error of the modulation.

[0144] S8: Estimate the residual phase error of the azimuth based on the MEA;

[0145] In step S8, the estimation of the residual phase error of the azimuth is carried out by the model of the phase error in step S3.2 based on the minimum entropy criterion. After l iterations, it is judged whether the image entropy value Ent obtained satisfies formula (13). If it satisfies, the residual phase error of the azimuth that needs to be compensated is:

[0146]

[0147] S9: Perform fine compensation of the azimuth based on the estimated residual phase error;

[0148] In step S9, according to the residual phase error estimated in step S8 The compensation function of the residual phase error of the azimuth is and multiplied by formula (21) to perform fine compensation of the azimuth of the echo signal, and the following can be obtained:

[0149]

[0150] S10: Perform azimuth compression on the echo signal after fine compensation of the azimuth to obtain a THz-SAR focusing image;

[0151] In step S10, the echo signal after fine compensation of the azimuth is matched filtered, and after azimuth compression, the model expression of the THz-SAR focusing image obtained is:

[0152]

[0153] Wherein: B d is the Doppler bandwidth.

[0154] In summary, in the embodiment, the proposed THz SAR two-dimensional self-focusing imaging algorithm comprehensively considers the nonlinear phase error in the range direction and the motion phase error in the azimuth direction, and estimates and compensates accordingly to obtain a high-resolution THz-SAR focusing image. At the same time, based on the joint use of IMU / GPS measurement data and MEA self-focusing algorithm, compared with directly using the self-focusing algorithm, the computational complexity is effectively reduced.

[0155] ​The above description is merely a preferred embodiment of the present invention and does not limit the present invention in any way. Any person skilled in the art who, without departing from the scope of the present invention, makes any equivalent substitution, modification, or other changes to the technical solution and technical content disclosed in the present invention shall be deemed to be within the scope of the present invention and still fall within the scope of protection of the present invention.

Claims

1. A terahertz SAR two-dimensional self-focusing imaging algorithm, characterized in that: The following steps are involved: S1: Establish a model of the echo signal of terahertz SAR; S2: removing the remaining video phase term and the oblique phase term in the echo signal; S3: Estimate the nonlinear phase error in the range direction; S4: performing range compensation based on the estimated nonlinear phase error; S5: performing distance compression on the echo signal after distance compensation; S6: Based on the IMU / GPS measurement data, perform coarse azimuth compensation on the echo signal after the range compression; S7: performing range migration correction on the echo signal after coarse azimuth compensation; S8: Estimate the residual phase error of the azimuth direction; S9: performing azimuth precise compensation based on the estimated residual phase error; S10: performing azimuth compression on the echo signal after azimuth precise compensation to obtain a THz-SAR focused image; In step S3, it is assumed that the function of the nonlinear phase error in the range direction is Then the model expression of the echo signal with the nonlinear phase error is: Where: τ is the fast time in distance; t is the slow time in azimuth; Discretize τ and t into t n (n=0:N-1) and t m (m=0:M-1), then the discretization of formula (6) can be obtained: The minimum entropy criterion is used as the evaluation criterion, and the nonlinear phase error is estimated through iterative optimization. The specific steps are as follows: S3.1: Assume that the estimated nonlinear phase error in the range direction is and The initial setting is 0, that is, the range image expression after compensation is: Where: k is the range frequency; m is the echo pulse; n is the range unit; N is the total number of range units; M is the total number of azimuth units; S3.2: Based on the minimum entropy criterion, a phase error model is established, which is expressed as: Where: Ent is the image entropy value, the expression is: Where: E is the range image energy, the expression is: S3.3: Solving equation (9) based on Newton's method, the iterative expression of the phase error can be obtained as follows: in: Where: s0 (l-1) (n, m) is the echo data corrected using the phase error estimated by the (l-1)th iteration; G k，m Take the complex conjugate of each element in; S3.4: Determine whether the estimated phase error is accurate enough, that is, determine the image entropy values ​​Ent and Ent obtained after the lth and (l-1th) iterations respectively. (l-1) Is the difference between them lower than the preset threshold value J, that is: |Ent (l) -Ent (l-1) |≤J (13) If equation (13) is satisfied, the nonlinear phase error in the range direction that needs to be compensated is: In step S4, the nonlinear phase error estimated in step S3 is Then the compensation function of the nonlinear phase error in the range direction is And multiply it by formula (7) to compensate the echo signal in the range direction, and we can get: In step S7, the echo signal after the coarse azimuth compensation is subjected to an azimuth Fourier transform (FFT), the echo signal is transformed into the range Doppler domain, and the echo signal is corrected for range migration by interpolation. After eliminating the influence of the range migration, the echo signal is subjected to an azimuth inverse Fourier transform (IFFT), and the following is obtained: in: is the discrete delay time of the shortest slant range from the target to the radar; B is the range signal bandwidth; residual phase error In step S8, based on the minimum entropy criterion, the residual phase error in the azimuth direction is estimated using the phase error model in step S3.

2. After 1 iteration, it is determined whether the obtained image entropy value Ent satisfies equation (13). If so, the residual phase error in the azimuth direction that needs to be compensated is: In step S9, the residual phase error estimated in step S8 is Then the compensation function of the residual phase error in azimuth is And multiply it by formula (21) to perform azimuth precise compensation on the echo signal, and we can get: Where: R Δ is the difference between the actual distance of the radar and the reference distance; is the discrete slow time for the beam center to pass through the target; w a (·) is the azimuthal envelope.

2. The terahertz SAR two-dimensional self-focusing imaging algorithm according to claim 1, characterized in that: In step S1, the terahertz SAR transmits a linear frequency modulated pulse signal and performs linear frequency demodulation on the received echo signal. The model expression of the generated echo signal is: Where: τ is the fast time in distance; t is the slow time in azimuth; T p is the pulse width; c is the speed of light; λ is the wavelength; R i is the actual distance of the radar; R ref is the reference distance of the radar; j is the imaginary unit; γ is the frequency modulation slope.

3. The terahertz SAR two-dimensional self-focusing imaging algorithm according to claim 2, characterized in that: In step S2, the range-direction Fourier transform (FFT) of equation (1) is performed to obtain: Where: f r is the distance frequency; f c is the center frequency; is the Doppler term; is the residual video phase term; is the oblique phase term of the echo envelope; The residual video phase term and the oblique phase term in equation (2) are expressed as: Formula (2) multiplied by the compensation function of the residual video phase term and the oblique phase term We can get: Performing an inverse fast Fourier transform (IFFT) on equation (4), the model expression of the echo signal after the transformation is:

4. The terahertz SAR two-dimensional self-focusing imaging algorithm according to claim 1, characterized in that: In step S5, the echo signal after range compensation is subjected to range Fourier transform (FFT), and the model expression of the echo signal after range compression is: Where: t i is the discrete delay time from target to radar; Based on motion phase error The existence of , in fact, the model expression of the echo signal that completes the distance compression is:

5. The terahertz SAR two-dimensional self-focusing imaging algorithm according to claim 4, characterized in that: In step S6, based on the IMU / GPS measurement data, the difference d between the actual slant distance between the antenna and the scene and the ideal slant distance is calculated. los (t n ,t m ), in order to perform corresponding motion phase compensation, and the compensation function of the motion phase error is: Multiplying equation (17) by equation (18) to perform coarse azimuth compensation on the echo signal yields: Based on the residual phase error in the azimuth due to the IMU / GPS measurement data By transforming formula (19), we can get:

6. The terahertz SAR two-dimensional self-focusing imaging algorithm according to claim 1, characterized in that: In step S10, matched filtering is performed on the echo signal after fine compensation. After azimuth compression is completed, the model expression of the obtained THz-SAR focused image is: Among them: B d is the Doppler bandwidth.

Images