An Ellipsoidal Model-Based Near-Range Wide-Area One-Station Fixed FENLCS Imaging Method and System

Abstract

This invention discloses a near-range wide-area single-station fixed FENLCS imaging method and system based on an ellipsoidal model. The method steps are as follows: S11. Construct a geometric model for near-range wide-area single-station fixed bistatic SAR imaging and analyze its echo characteristics and spatially variable characteristics of the oblique angle; S12. Perform range-direction preprocessing on the target echo signal in the range frequency domain; S13. Based on the results of the range-direction preprocessing, construct an ellipsoidal model to obtain the azimuth spatially variable relationship between the receiver center slant range and the oblique angle of the target at any azimuth time point within the same range cell, and improve the range-direction processing accuracy using the residual higher-order RCMC method of azimuth spatial variation; S14. Remodel the azimuth spatially variable Doppler parameters based on the constructed ellipsoidal model, and use the FENLCS algorithm to achieve azimuth equalization and compression to obtain the final focused image. This invention improves the performance of SAR imaging.

Description

Technical Field

[0001] This invention belongs to the field of signal processing technology, and particularly relates to a near-range wide-area one-station fixed FENLCS imaging method and system based on an ellipsoidal model. Background Technology

[0002] Synthetic Aperture Radar (SAR) is a radar system that uses the relative motion between the radar and the target to synthesize a larger equivalent antenna aperture from a smaller real antenna aperture through signal processing. SAR features high resolution, all-weather operation, and the ability to penetrate cover, and is therefore often combined with mobile platforms such as UAVs and helicopters, finding wide application in both military and civilian fields.

[0003] A fixed-station synthetic aperture radar (SAR) is a special bistatic SAR geometry model with a fixed transmitter. Because the relative position between the transmitter and receiver changes with azimuth and time, the echo exhibits spatially variable azimuth characteristics, making imaging processing more complex. Furthermore, in a close-range, wide-field configuration, the echo characteristics of a mobile platform are more complex than in a typical long-range configuration. The traditional receiver slant angle invariance no longer holds, and the two-dimensional spatially variable slant angle exacerbates the range and azimuth spatial variations of the echo signal, further increasing the difficulty of SAR imaging.

[0004] Analysis of a near-range, wide-field, single-station fixed SAR imaging model reveals that signal echo characteristics are primarily influenced by the spatially variable receiver slant angle, making range migration correction in the range direction and Doppler parameter equalization in the azimuth direction difficult. Therefore, this invention proposes a near-range, wide-field, single-station fixed geometric model. An ellipsoidal analytical model is established to more accurately describe the azimuth spatially variable characteristics of the receiver slant angle and receiver slant range within the same range cell. Furthermore, a new azimuth frequency-domain extended non-linear chirp scaling (FENLCS) algorithm is derived to complete the azimuth equalization process and obtain the focusing result. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes a near-range wide-field one-station fixed FENLCS imaging method and system based on an ellipsoidal model. This invention obtains an analytical expression for the spatial variation relationship of point targets under the near-range wide-field one-station fixed configuration, and applies it to the subsequent processing of residual RCMC of azimuth spatial variation and azimuth Doppler parameter equalization, thereby improving the performance of SAR imaging.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A near-range, wide-area, one-station fixed FENLCS imaging method based on an ellipsoidal model includes:

[0008] S1. Construct a geometric model for short-range wide-area single-station fixed bistatic SAR imaging and analyze its echo characteristics and spatially variable characteristics of oblique angle.

[0009] S2. Perform range-direction preprocessing on the target echo signal in the range frequency domain;

[0010] S3. Based on the results of range preprocessing, an ellipsoidal model is constructed to obtain the azimuth spatial variation relationship of the receiver center slant range and slant angle of the target at any azimuth time point within the same range cell, and the residual high-order RCMC method of azimuth spatial variation is used to improve the processing accuracy of range.

[0011] S4. Based on the constructed ellipsoidal model, the azimuth spatial variation Doppler parameters are remodeled, and the FENLCS algorithm is used to achieve azimuth equalization and compression to obtain the final focused image.

[0012] Preferably, step S11 is as follows:

[0013] The transmitter is located at (x T The receiver is stationary at (0, 0, h), and flies along the X-axis at a velocity v. When the azimuth time t = 0, the receiver's position coordinates are (0, 0, h), and the corresponding r is... T0 r represents the distance from the transmitter to the scene center point P0(x0,y0,0) along the direction of the radar beam center. R0 Let r represent the distance from the receiver to point P0(x0,y0,0), and let θ0 represent its oblique angle. Tc It is t=t c The slant range r from the transmitter to the target P(x,y,0) at any given time. Rc It is t=t c The slant range from the receiver to the target P(x,y,0) at any given time; the angle β between the beam projection on the ground and the azimuth direction is a fixed azimuth angle, r g0 With r g Representing the zero time of azimuth and t respectively c The projection of the slant distance onto the ground at any given time; the spatial slant angle θ is represented as follows:

[0014]

[0015] Based on geometric relationships, the instantaneous transmit and receive slant range of target P can be expressed as:

[0016]

[0017] Based on this, at t=t c Performing four Taylor expansions at this point yields:

[0018] R total (t;r Rc ,t c )=Z+A(tt c )+B(tt c ) 2 +C(tt c ) 3 +D(tt c ) 4 (3)

[0019] in,

[0020]

[0021] In equation (4), A is the coefficient of the linear distance migration term, B is the coefficient of the quadratic distance migration term, and C and D are the coefficients of the higher-order distance migration term;

[0022] Based on the simulation parameters, the effects were analyzed using equations (1) and (5) respectively.

[0023]

[0024] Subsequently, the influence of the oblique angle of the air variable on LRCM and QRCM was quantitatively analyzed through simulation. The errors of LRCM and QRCM are defined as follows:

[0025]

[0026] Preferably, step S12 is as follows:

[0027] Assuming the radar transmitter sends a linear frequency modulated signal, the demodulated form of the echo signal for a point target P(x,y,0) is as follows:

[0028]

[0029] Among them, w r (·) represents the range envelope, w a (·) represents the azimuth envelope, c is the speed of light, and f is the speed of light. c τ is the signal carrier frequency, K is the range-direction time, and K is the range carrier frequency. r It is distance-based frequency modulation;

[0030] Equation (7) can be transformed to the range frequency domain using series inversion:

[0031]

[0032] Among them, f r It is the range frequency; the dominant factor in range-azimuth coupling is the linear range migration of the first term, which is processed using LRCMC, and the phase compensation expression is:

[0033]

[0034] Where A0=-vsin(θ0) represents the linear distance migration coefficient at the central reference point; by multiplying equation (9) and equation (8), we get:

[0035]

[0036] Resampling is performed using the KT transform, i.e., let t = t m f c / (f c +f r After resampling, press f r Taylor series expansion yields:

[0037]

[0038] Where φ0 is the azimuth modulation term, φ1 is the range-position term coefficient, and φ2 and φ3 are higher-order range-azimuth coupling terms, expressed as:

[0039]

[0040] Observational expression (11) reveals t m with f r The linear coupling terms have been completely eliminated; subsequently, the higher-order RCM terms are compensated, i.e., Bulk RCMC and quadratic distance compression are performed, and the compensated phase is expressed as:

[0041]

[0042] Where B0, C0, and D0 represent the distance migration coefficients at the central reference point, multiplying equation (13) by equation (11) yields the distance delay curve:

[0043]

[0044] in,

[0045]

[0046] In the above formula, μ0 represents the new point target position after Bulk RCMC correction, and Δμ(t) m ;r Rc ,t c ) indicates the residual higher-order RCM error.

[0047] Preferably, step S13 is as follows:

[0048] Suppose P is a selected imaging point target in the scene region. After distance preprocessing, the distance it represents is shifted towards the focusing position, as shown by Z = r Tc +r Rc Offset is R total (0;r Rc ,t c ); where R total (0;r c ,t c ) represents the sum of the transmit and receive slant ranges of the point target at azimuth zero time, denoted by TP+RP; if the point target P is in the same range cell as the reference point P0 after range shift, then TP+RP = TP0+RP0 is considered. Furthermore, a special ellipsoid is constructed with the receiver and transmitter positions at azimuth zero time as the two foci, and the semi-axis lengths along the Y-axis and Z-axis are equal. The point targets P0 and P are considered to be on the elliptical surface where the ellipsoid intersects with the XOY plane.

[0049] Therefore, the equation of the ellipsoid is:

[0050]

[0051] Where, m=(r T0 +r R0 ) / 2, n=TR / 2; and based on this, the equation of the intersecting ellipse is obtained as:

[0052]

[0053] Among them, the semi-major axis distance a and the semi-focal distance c of the ellipse are derived from equation (16), and the eccentricity e of the ellipse is equal to c / a.

[0054] By using the geometric relationships in the ellipsoidal analytical model, the azimuth spatial variation relationship of the receiving oblique angle is derived as follows:

[0055]

[0056] The linear term coefficients U1 and V1 are as follows:

[0057]

[0058] The relationship between the receiving slant range and the azimuth spatial variation is obtained by solving:

[0059]

[0060] Specifically, the coefficients Z1 and Z2 of the linear term are:

[0061]

[0062] If we take e = 0, h = 0, a = rR0 β=π / 2-θ0, then the spatial variation relationship of the receiving slant range becomes r Rc =r R0 -vsin(θ0)t c ;

[0063] The instantaneous transmit and receive slant ranges and the expansion coefficients A, B, C, and D of equation (3) all vary with the receiving slant range r. Rc And considering the change in the angle of view θ, we substitute equations (18) and (20) into the expansion coefficients for remodeling. The Taylor expansion is:

[0064]

[0065] in,

[0066]

[0067] To achieve residual higher-order RCM correction for azimuth spatial variation, substituting equation (23) into equation (14) yields:

[0068]

[0069] Equation (24) shows that the remaining higher-order RCM has been converted into the representation of azimuth spatial variation components, and its correction method is achieved by constructing the following compensation filter:

[0070]

[0071] Multiplying equation (24) by equation (25), at t m =t c Performing a Taylor series expansion at this point, and setting the coefficients of the linear terms to zero, we obtain:

[0072]

[0073] Preferably, step S14 is as follows:

[0074] The azimuth modulation term in equation (11) is placed at t m =t c Perform a Taylor series expansion at this point and rewrite it as:

[0075]

[0076] in,

[0077]

[0078] In equation (27), λ = c / f c f d1 -f d10 f is the residual Doppler center frequency. d2For azimuth tuning frequency, f d3 f d4 The coefficients of higher-order terms are represented by the azimuth spatial variation relation derived from the ellipsoidal analytical model. The coefficients are then remodeled using the following expression:

[0079]

[0080] Subsequently, the signal is converted to the azimuth frequency domain and its constant term is ignored, resulting in:

[0081]

[0082] Expanding the coefficients in equation (30) into non-azimuth spatial variation parts and azimuth spatial variation parts, we get:

[0083]

[0084] Next, we introduce a high-order non-empty variable pre-filtering process in the frequency domain, which is to multiply equation (30) by the following filter:

[0085]

[0086] To mitigate the spatial variation effect of azimuth, the pre-filtered signal is converted to the azimuth time domain and multiplied by the following time-domain phase adjustment factor:

[0087]

[0088] Where Y3 and Y4 are undetermined coefficients; the processed result is subjected to Fourier transform using MSR to convert the signal to the azimuth frequency domain, and then multiplied by the FENLCS factor:

[0089]

[0090] Where p2, p3, and p4 are all undetermined coefficients; transforming the multiplied signal to the azimuth time domain yields:

[0091]

[0092] In equation (35), A Ω It is a directional compression term, B Ω The focus position information of the point target is included, while the other terms are coupling terms that reduce the azimuth focusing quality; to eliminate the influence of Doppler spatial variation parameters, the first-order coupling term B is... Ω Set to -f d20 π / ε, the coefficients of higher-order coupling terms are then 0, that is:

[0093]

[0094] In order to ensure that the focal position of the point target is consistent with its actual position, the value of ε should be as close as possible to 0.5; then, solving equation (36) yields:

[0095]

[0096] Substituting the above coefficients into A Ω In the process, the final azimuth compression factor is obtained, expressed as:

[0097]

[0098] Substituting equation (36) into equation (35) and multiplying by equation (38) completes the azimuth focusing process.

[0099] This invention also discloses a near-range wide-area one-station fixed FENLCS imaging system based on an ellipsoidal model, comprising the following modules connected in sequence:

[0100] The spatial variation characteristics analysis module for echo characteristics and oblique angle: Constructs a geometric model of a near-range wide-field one-station fixed bistatic SAR imaging system and analyzes its echo characteristics and oblique angle spatial variation characteristics.

[0101] Range preprocessing module: performs range preprocessing on the target echo signal in the range frequency domain;

[0102] Range processing accuracy improvement module: Based on the results of range preprocessing, an ellipsoidal model is constructed to obtain the azimuth spatial variation relationship of the receiver center slant range and slant angle of the target at any azimuth time point within the same range cell, and the residual high-order RCMC method of azimuth spatial variation is used to improve the processing accuracy of range.

[0103] The final focused image acquisition module: Based on the constructed ellipsoidal model, the azimuth spatial variation Doppler parameters are remodeled, and the FENLCS algorithm is used to achieve azimuth equalization and compression to obtain the final focused image.

[0104] The specific processing flow of this invention is as follows: First, linear range cell migration correction (LRCMC) and range compression are performed on the SAR echo signal in the range direction to remove most of the linear range cell migration (RCM). Then, the Keystone Transform (KT) is used to eliminate residual linear RCM and alleviate some range-azimuth coupling. Next, the corrected signal form is obtained through Bulk Range Cell Migration Correction (Bulk RCMC), and an ellipsoidal analytical model is constructed based on the signal characteristics to accurately describe the spatial dependence of the receiving slant range and receiving angle of point targets within the same range cell after correction. Then, the azimuth spatial variation relationship obtained from this model is applied to the processing of residual higher-order RCMC. Similarly, the azimuth spatial variation relationship is applied to the FENLCS algorithm in the azimuth direction to achieve Doppler center frequency removal, Doppler frequency modulation equalization, and azimuth compression, completing the final focusing.

[0105] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0106] This invention employs LRCMC processing to remove the main linear RCM. Next, KT transform is used to remove the residual linear RCM term of the point target at azimuth zero time, alleviating some range-azimuth coupling. Combining the signal echo after Bulk RCMC processing, this invention proposes an ellipsoidal analytical model. By applying the spatial position analytical expression of the point target within the same range cell derived from this model to range residual high-order RCM correction processing and azimuth FENLCS algorithm processing, the accuracy of range processing is improved, spatially variable Doppler center frequency removal and frequency modulation equalization are achieved, and finally, azimuth focusing is completed. Simulation results demonstrate the effectiveness of the ellipsoidal analytical model and corresponding signal processing algorithm constructed in this invention under close-range, wide-area, single-station fixed imaging conditions. Attached Figure Description

[0107] Figure 1 This is a flowchart of the near-range wide-area one-station fixed FENLCS imaging method based on an ellipsoidal model provided in Example 1;

[0108] Figure 2 The algorithm processing flowchart provided in Example 1;

[0109] Figure 3 These are schematic diagrams of the geometric configurations of a near-range, wide-area, single-station fixed SAR imaging system provided in Examples 1 and 2.

[0110] Figure 4A schematic diagram of the oblique angle spatial variation characteristics and echo characteristics provided in Example 1 ( Figure 4 (a) is a schematic diagram showing the change of the receiver's oblique angle with ground distance in Embodiment 1; Figure 4 (b) is a schematic diagram showing the variation of slant range error with ground distance provided in Example 1;

[0111] Figure 5 A schematic diagram of distance migration error provided for Example 1 ( Figure 5 (a) is a schematic diagram of linear distance migration error provided in Example 1; Figure 5 (b) is a schematic diagram of the secondary distance migration error provided in Example 1;

[0112] Figure 6 This is a schematic diagram of the distance migration trajectory after Bulk RCMC processing provided in Example 1;

[0113] Figure 7 A schematic diagram of the analytical model of the ellipsoid provided in Example 1;

[0114] Figure 8 This is a schematic diagram of the ellipsoidal analytical model provided in Example 1 projected onto the XOY plane.

[0115] Figure 9 This is a schematic diagram of the distance migration trajectory after residual high-order RCMC processing provided in Example 1;

[0116] Figure 10 This is a schematic diagram comparing the secondary phase error QPE provided in Example 2;

[0117] Figure 11 This is a schematic diagram comparing the actual focusing effect after processing, provided in Example 2. Figure 11 (a) is a schematic diagram of the focused image after processing by the traditional method; Figure 11 (b) is a schematic diagram of the focused image after processing by the method provided in Example 2;

[0118] Figure 12 A schematic diagram of the azimuth pulse compression profile of edge point P1 provided in Example 2 ( Figure 12 (a) is a schematic diagram of the azimuth pulse pressure profile of P1 after processing by the traditional method; Figure 12 (b) Schematic diagram of the azimuth pulse pressure profile of P1 after processing by the method provided in Example 2.

[0119] Figure 13 This is a block diagram of a near-range wide-area one-station fixed FENLCS imaging system based on an ellipsoidal model, provided in Embodiment 3. Detailed Implementation

[0120] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0121] To address the shortcomings of existing technologies, this invention proposes a near-range, wide-area, one-station fixed FENLCS imaging method based on an ellipsoidal model.

[0122] This invention performs LRCMC and range compression on the SAR echo signal in the range direction, removing most of the linear RCM. Subsequently, KT transform is used to eliminate residual linear RCM and alleviate some range-azimuth coupling. The corrected signal form is then obtained through BulkRCMC, and an ellipsoidal analytical model is constructed based on the signal characteristics to accurately describe the spatial dependence of the receiving slant range and receiving angle of point targets within the same range cell after correction. The azimuth spatial variation relationship obtained from this model is then applied to the processing of residual high-order RCMC. Similarly, the azimuth spatial variation relationship is applied to the FENLCS algorithm in the azimuth direction, achieving Doppler center frequency removal, Doppler frequency modulation equalization, and azimuth compression, thus completing the final focusing.

[0123] Example 1

[0124] like Figure 1-2 As shown, this embodiment provides a near-range, wide-area, one-station fixed FENLCS imaging method based on an ellipsoidal model, including the following:

[0125] S11. Construct a geometric model for near-range wide-area single-station fixed bistatic SAR imaging and analyze its echo characteristics and spatially variable characteristics of oblique angle.

[0126] S12. Perform range-direction preprocessing on the target echo signal in the range frequency domain;

[0127] S13. Based on the results of range preprocessing, an ellipsoidal model is constructed to obtain the azimuth spatial variation relationship of the receiver center slant range and slant angle of the target at any azimuth time point within the same range cell, and the residual high-order RCMC method of azimuth spatial variation is used to improve the processing accuracy of range.

[0128] S14. Based on the constructed ellipsoidal model, the azimuth spatial variation Doppler parameters are remodeled, and the FENLCS algorithm is used to achieve azimuth equalization and compression to obtain the final focused image.

[0129] In step S11 of this embodiment, the geometric configuration of a near-range wide-area one-station fixed SAR imaging system is constructed, and the echo characteristics and spatially variable characteristics of the geometric configuration are analyzed. Specifically, this includes:

[0130] like Figure 3 As shown, the transmitter is located at (x T ,y T The receiver is stationary at (0,0,h), and flies along the X-axis at a velocity v. When the azimuth time t = 0, the receiver's position coordinates are (0,0,h), and the corresponding r is... T0 r represents the distance from the transmitter to the scene center point P0(x0,y0,0) along the direction of the radar beam center. R0 Let represent the distance from the receiver to point P0(x0,y0,0), and let θ0 represent its oblique angle. Similarly, r Tc It is t=t c The slant range r from the transmitter to the target P(x,y,0) at any given time. Rc It is t=t c The slant range from the receiver to the target P(x,y,0) at any given time. Furthermore, the angle β between the beam projection onto the ground and the azimuth direction is a fixed azimuth angle, r. g0 With r g Representing the zero time of azimuth and t respectively c The projected length of the slant distance onto the ground at any given time. The spatial slant angle θ can be expressed as follows:

[0131]

[0132] Based on geometric relationships, the instantaneous transmit and receive slant range of target P can be expressed as:

[0133]

[0134] Based on this, at t=t c Performing four Taylor expansions at this point yields:

[0135] R total (t;r Rc ,t c )=Z+A(tt c )+B(tt c ) 2 +C(tt c ) 3 +D(tt c ) 4 (3)

[0136] in,

[0137]

[0138] In equation (4), A is the coefficient of the Linear Range Cell Migration (LRCM) term, B is the coefficient of the Quadratic Range Cell Migration (QRCM) term, and C and D are the coefficients of the higher-order range migration terms. It can be seen that all RCM coefficients are related to the receiver slant range r. Rc The correlation with the receiver's angle of view θ further proves that its SAR echo characteristics have complex two-dimensional spatial variation characteristics.

[0139] Table 1

[0140] Simulation parameters Simulation values Simulation parameters Simulation values carrier frequency 10GHz transmitter height 3km Distance bandwidth 150MHz Receiver height 3km Doppler bandwidth 170Hz Receiver speed 170m / s Synthetic pore size time 2.45s Reference ground distance 10km Pulse repetition frequency 536Hz Reference point receives oblique angle 52°

[0141] To further analyze the receiver's oblique angle θ and receiver oblique distance r Rc Based on the distance dependence characteristics, this embodiment uses the simulation parameters shown in Table 1 to simulate and analyze the influence of equations (1) and (5) on the distance dependence characteristics. The results are as follows: Figure 4 As shown.

[0142]

[0143] from Figure 4 As shown in (a), when the ground distance is in the long range (≥30km), the receiver's angle of view tends to remain constant, and traditional imaging algorithms can achieve ideal imaging results. However, when the ground distance is in the short range (≤15km), the receiver's angle of view changes drastically. Therefore, neglecting the spatial variation of the angle of view will significantly impact the imaging effect. Furthermore, from... Figure 4 As shown in (b), the Slant Range Error (SRE) is still relatively large in the short range, which will lead to serious distance migration error and cannot be ignored.

[0144] Subsequently, this embodiment quantitatively analyzes the influence of the oblique angle of the air variable on LRCM and QRCM through simulation. The influence of higher-order terms is very small and therefore can be ignored. The errors of LRCM and QRCM are defined as follows:

[0145]

[0146] like Figure 5 As shown in (a), the linear range migration error is less than half the range resolution only within 200m of the reference point, indicating that if the slant angle effect of spatial variation is not considered, linear range migration correction is only applicable to a very small range region. Therefore, the residual LRCM after LRCMC requires further processing. Figure 5 As shown in (b), the secondary distance migration error is small and its impact on imaging can be ignored.

[0147] In step S12, the target echo signal is preprocessed in the range direction.

[0148] In this embodiment, the echo signal was processed by LRCMC to correct the non-space-varying linear RCM, and the linear RCM at azimuth zero time was completely eliminated by KT transform. Subsequently, higher-order RCM correction was achieved using Bulk RCMC, specifically including:

[0149] First, assuming the radar transmitter is sending a linear frequency modulated signal (LFM), the demodulated form of the echo signal for a point target P(x,y,0) is as follows:

[0150]

[0151] Among them, w r (·) represents the range envelope, w a (·) represents the azimuth envelope, c is the speed of light, and f is the speed of light. c τ is the signal carrier frequency, K is the range-direction time, and K is the range carrier frequency. r It is distance-based frequency modulation.

[0152] Equation (7) is transformed to the range frequency domain using the Method of Series Reverse (MSR):

[0153]

[0154] Where f r This refers to the range frequency. Based on the above analysis, the dominant factor in range-azimuth coupling is the linear range migration of the first term. Therefore, LRCMC is used for processing, and its phase compensation expression is:

[0155]

[0156] Where A0 = -vsin(θ0) represents the linear distance migration coefficient at the central reference point. The product of equation (9) and equation (8) yields:

[0157]

[0158] While the above processing can mitigate the impact of linear distance migration, the large error in the LRCM term caused by two-dimensional spatial variation in the near-range wide-domain environment cannot be completely eliminated by reference point phase compensation. Therefore, residual terms (A-A0)(tt) still exist, affecting imaging processing. c ).

[0159] To completely eliminate the residual linear components of the two-dimensional spatial variation, resampling can be performed using the KT transform, i.e., let t = t m f c / (f c +f r After resampling, press f r Taylor series expansion yields:

[0160]

[0161] Where φ0 is the azimuth modulation term, φ1 is the range-position term coefficient, and φ2 and φ3 are higher-order range-azimuth coupling terms, expressed as:

[0162]

[0163] Observing equation (11), we can find that t m with f r The linear coupling terms have been completely eliminated. Subsequently, compensation is performed on its higher-order RCM terms, namely, Bulk RCMC and Second Range Compression.

[0164] Compression (SRC) processing, its compensated phase can be expressed as:

[0165]

[0166] Where B0, C0, and D0 represent the distance migration coefficients at the central reference point, multiplying equation (13) by equation (11) yields the distance delay curve:

[0167]

[0168] in,

[0169]

[0170] In the above formula, μ0 represents the new point target position after Bulk RCMC correction, and Δμ(t) m ;r Rc ,t c ) represents the residual higher-order RCM error. Analysis of equation (15) reveals that, through the aforementioned imaging processing, the actual focal position of the point target changes from the original Z = r Tc +r Rc It became R total (0,r Rc ,t c That is, point targets with the same transmit and receive slant range will be focused on the same range unit after range processing.

[0171] Furthermore, this invention obtained schematic diagrams of the aforementioned range-processed RCM trajectories through simulation. For example... Figure 6 As shown, due to the presence of residual higher-order RCM errors in azimuth spatial variation, the RCM trajectory of edge point targets within the same range cell cannot be considered as a straight line that has been corrected to be parallel to the azimuth direction. Subsequent azimuth focusing operations are difficult to perform. Therefore, obtaining an accurate azimuth spatial variation relation to eliminate residual higher-order RCM terms is very important for high-precision imaging in close-range wide-field conditions.

[0172] In step S13, based on the results of range preprocessing, an ellipsoidal model is constructed to obtain the azimuth spatial variation relationship of the receiver center slant range and slant angle of the target at any azimuth time point within the same range cell. The residual high-order RCMC method of azimuth spatial variation is used to improve the processing accuracy of the range direction.

[0173] In this embodiment, an ellipsoidal model was constructed, and an analytical expression was derived that can more accurately describe the azimuth spatial variation relationship of the received slant range and received slant angle of point targets located in the same range cell after range preprocessing. This expression was then applied to the residual higher-order RCMC processing of azimuth spatial variation, thereby improving the imaging accuracy in the range direction. The specific steps are as follows:

[0174] like Figure 7 As shown, assuming P is a selected imaging point target in the scene area, after the distance preprocessing described above, the distance it represents shifts towards the focusing position. Based on the above analysis, this is due to Z = r Tc +r Rc Offset is R total (0;r Rc ,t c ). Among them, R total (0;r c ,t c The sum of the transmit and receive slant ranges of the point target at azimuth zero time can be represented by TP + RP. Therefore, if point target P is within the same range cell as reference point P0 after range shifting, then TP + RP = TP0 + RP0 can be considered. Further, a special ellipsoid is constructed with the receiver and transmitter positions at azimuth zero time as two foci, and the semi-axis lengths along the Y and Z axes being equal. Point targets P0 and P can then be considered to lie on the elliptical surface where this ellipsoid intersects with the XOY plane, as illustrated in the diagram below. Figure 8 As shown.

[0175] Therefore, the equation of the ellipsoid can be obtained as follows:

[0176]

[0177] Where, m=(r T0 +r R0) / 2, n=TR / 2. And based on this, we can obtain Figure 8 The equations of the intersecting ellipses shown are:

[0178]

[0179] The semi-major axis distance a and semi-focal distance c of the ellipse can be derived from equation (16), and the eccentricity e of the ellipse is equal to c / a.

[0180] Based on the preceding text, by using the geometric relationships in the ellipsoidal analytical model to solve the problem, the azimuth spatial variation relationship for the received oblique angle can be derived as follows:

[0181]

[0182] The linear term coefficients U1 and V1 are as follows:

[0183]

[0184] Similarly, the relationship between the receiving slant range and the azimuth variation can also be solved:

[0185]

[0186] The coefficients Z1 and Z2 of the linear term are as follows:

[0187]

[0188] If we take e = 0, h = 0, a = r R0 β=π / 2-θ0, then the spatial variation relationship of the receiving slant range becomes r Rc =r R0 -vsin(θ0)t c The expression obtained is the same as that obtained by the traditional model, which also verifies the correctness of the ellipsoid model constructed in this invention.

[0189] Based on the previous analysis, the expansion coefficients A, B, C, and D of the instantaneous transmit and receive slant ranges and (3) all vary with the receiving slant range r. Rc And considering the change in the angle of view θ, we substitute equations (18) and (20) into the expansion coefficients for remodeling, and the Taylor expansion is:

[0190]

[0191] in,

[0192]

[0193] Next, in order to achieve residual higher-order RCM correction for azimuth spatial variation, substituting equation (23) into (14) yields:

[0194]

[0195] Observing formula (24), we can see that the remaining higher-order RCM has been transformed into the representation of azimuth spatial variation components, and its correction method can be achieved by constructing the following compensation filter:

[0196]

[0197] Multiplying equations (24) and (25), at t m =t c Performing a Taylor series expansion at this point, and setting the coefficients of the linear terms to zero, we obtain:

[0198]

[0199] To further verify the effect of residual higher-order RCM correction, simulations were performed according to the parameters in Table 1, and the results are as follows: Figure 9 As shown, it can be observed that after the above series of processing, the distance migration error of the azimuth edge point target is much less than half of the range resolution unit, which also ensures the imaging accuracy in the case of close range and wide field.

[0200] In step S14, the Doppler parameters of azimuth spatial variation are remodeled based on the constructed ellipsoidal model, and the FENLCS algorithm is used to achieve azimuth equalization and compression to obtain the final focused image.

[0201] In this embodiment, the azimuth spatial variation relation derived from the ellipsoidal analytical model is applied to the remodeling of Doppler parameters and then to the FENLCS algorithm to achieve azimuth equalization and compression, specifically including:

[0202] As can be seen from the above analysis, point targets located in the same range cell after range processing have different initial slant ranges and different Doppler parameters. Therefore, it is necessary to eliminate the azimuth spatial variation of the Doppler parameters before performing azimuth compression.

[0203] The azimuth modulation term in equation (11) is placed at t m =t c Perform a Taylor series expansion at this point and rewrite it as:

[0204]

[0205] in,

[0206]

[0207] In equation (27), λ = c / f c f d1 -f d10 f is the residual Doppler center frequency. d2For azimuth tuning frequency, f d3 f d4 This represents the coefficients of higher-order terms. Considering the two-dimensional spatial variation existing in the near-range wide-field case, the influence of the residual Doppler center frequency on imaging cannot be directly ignored. Therefore, this invention uses the azimuth spatial variation relation derived from the ellipsoidal analytical model to remodel the coefficients, and its expression is as follows:

[0208]

[0209] Subsequently, converting the signal to the azimuth frequency domain and ignoring its constant terms yields:

[0210]

[0211] Similarly, expanding the coefficients in equation (30) into non-azimuth spatial variation parts and azimuth spatial variation parts, we get:

[0212]

[0213] Next, we introduce a high-order non-empty variable pre-filtering process in the frequency domain, which is to multiply equation (30) by the following filter:

[0214]

[0215] To mitigate the spatial variation effect of azimuth, the pre-filtered signal is converted to the azimuth time domain and multiplied by the following time-domain phase adjustment factor:

[0216]

[0217] Where Y3 and Y4 are undetermined coefficients. The processed result is then subjected to a Fourier transform using MSR to convert the signal to the azimuth frequency domain, and multiplied by the FENLCS factor:

[0218]

[0219] Where p2, p3, and p4 are all undetermined coefficients. Converting the multiplied signal to the azimuth-time domain yields:

[0220]

[0221] Observational formula (35), A Ω It is a directional compression term, B Ω The focus position information for point targets is included, while the other terms are coupling terms that reduce the azimuth focusing quality. To eliminate the influence of Doppler spatial variation parameters, the first-order coupling term B is... Ω Set to -f d20 π / ε, the coefficients of higher-order coupling terms are then 0, that is:

[0222]

[0223] To ensure that the focused position and the actual position of the point target are consistent, the value of ε should be as close as possible to 0.5. Solving equation (36) then yields:

[0224]

[0225] Substituting the above coefficients into A Ω In the process, the final azimuth compression factor is obtained, which can be expressed as:

[0226]

[0227] Substituting equation (36) into equation (35) and multiplying by equation (38) will complete the azimuth focusing process.

[0228] Example 2:

[0229] like Figure 3 As shown, the transmitter is located at (x T ,y T The receiver is stationary at point R (h), and flies along the X-axis at a speed v. When the azimuth time t = 0, the receiver is located at point R, at altitude h, with its beam center illuminating the point target P0, and its receiving slant range r. R0 When t = t c At that time, the receiver is located at point M, and the corresponding receiving slant range is r. Rc Among them, r T0 With r Tc These represent the center slant distances from the transmitter to the point target at two different azimuth times.

[0230] Table 2 shows the system parameters used in this illustrative example, with the range and azimuth widths of the imaging area both being 2.0 km. The center of the scene is P0, where P1 and P2 are in the same range unit as P0, with an azimuth interval of 1.0 km; P3 and P4 are in the same azimuth position as P0, with a range interval of 1.0 km.

[0231] Table 2

[0232]

[0233]

[0234] To demonstrate the improved effect of the proposed method, the Doppler modulation frequency of different models was evaluated using the parameters listed in Table 2. The results of their quadratic phase error (QPE) curves are shown below. Figure 10 As shown.

[0235] from Figure 10It can be seen that, for the traditional model, due to insufficient modeling accuracy, the error exceeds the threshold ±π / 4 when the effective width of the imaging scene in the azimuth direction exceeds 500m. However, the QPE corresponding to the ellipsoidal analytical model can remain below ±π / 4 over a much larger range. This means that the model is suitable for short-range, wide-area, one-station fixed SAR. Furthermore, observation shows that at zero distance from the azimuth center, the QPE value of both the traditional model and the ellipsoidal analytical model of this invention is 0, indicating no processing error. Therefore, the main focus of this focusing algorithm is the focusing effect in the azimuth direction.

[0236] To observe the specific focusing effect on the point target, echo simulations were performed on the point target according to the parameters in Table 2. The simulations were then processed using imaging methods based on both the traditional model and the proposed ellipsoidal analytical model. The simulation results are shown below. Figure 11 As shown. Figure 11 (a) shows the results obtained by the traditional model algorithm. The focusing energy of the edge point target P1 is not concentrated, the focusing effect is poor, and its azimuth profile also shows obvious distortion, such as Figure 12 As shown in (a); and as Figure 11 (b) shows the results obtained by the ellipsoidal analytical model algorithm proposed in this invention. It can be observed that the focusing energy of the edge point target P1 is concentrated, and the energy is distributed from its azimuth to the profile. Figure 12 The Peak Side Lobe Ratio (PSLR) and Integrated Side Lobe Ratio (ISLR) calculated from the curves shown in (b) are close to the ideal values, meeting the imaging requirements. This also proves that the algorithm proposed in this invention greatly improves the imaging accuracy.

[0237] Example 3

[0238] like Figure 13 As shown, this embodiment of the near-range wide-area one-station fixed FENLCS imaging system based on an ellipsoidal model includes the following modules connected in sequence:

[0239] The spatial variation characteristics analysis module for echo characteristics and oblique angle: Constructs a geometric model of a near-range wide-field one-station fixed bistatic SAR imaging system and analyzes its echo characteristics and oblique angle spatial variation characteristics.

[0240] Range preprocessing module: performs range preprocessing on the target echo signal in the range frequency domain;

[0241] Range processing accuracy improvement module: Based on the results of range preprocessing, an ellipsoidal model is constructed to obtain the azimuth spatial variation relationship of the receiver center slant range and slant angle of the target at any azimuth time point within the same range cell, and the residual high-order RCMC method of azimuth spatial variation is used to improve the processing accuracy of range.

[0242] The final focused image acquisition module: Based on the constructed ellipsoidal model, the azimuth spatial variation Doppler parameters are remodeled, and the FENLCS algorithm is used to achieve azimuth equalization and compression to obtain the final focused image.

[0243] Other details in this embodiment can be found in Embodiment 1.

[0244] The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

Claims

1. A method for near field wide area one station fixed FENLCS imaging based on ellipsoid model, characterized in that According to the following steps: S1. Construct the geometric model of near-range wide-area one-station fixed dual-station SAR imaging, and analyze the space-varying characteristics of echo characteristics and squint angle; S2. Preprocess the target echo signal in the range frequency domain; S3. According to the result of the range direction preprocessing, construct an ellipsoid model to obtain the azimuth space-varying relationship of the receiver center slant range and squint angle of the target at any azimuth time point in the same range unit, and improve the processing accuracy of the range direction by using the residual high-order RCMC method of azimuth space variation; S4. Based on the constructed ellipsoid model, re-model the Doppler parameters of azimuth space variation, and realize the equalization and compression of the azimuth direction by using the FENLCS algorithm to obtain the final focused image; Step S3 is specifically as follows: Assume P is a selected imaging point target in the scene region, whose represented distance to the focus position is shifted by Z=r Tc +r Rc The shift is R total (0; r Rc , t c ); wherein, r Tc is t=t c the slant range of the transmitter to the target at time P ( x , y, 0), r Rc is t=t c the slant range of the receiver to the target at time P ( x , y, 0) ; R total (0; r c , t c ) denotes the slant range of the point target to the transmitter and the receiver at the time of bearing zero, and is denoted by TP + RP ; if the point target P is within the same distance unit as the reference point P 0 after the distance shift, then TP + RP=TP 0+ RP 0, further constructing a special ellipsoid with the positions of the receiver and the transmitter at the time of zero as two foci, and the half-axes along the Y axis and the Z axis being equal in length, the point target P 0 and P are considered to be on the intersecting elliptical surface of the ellipsoid and the XOY plane; Therefore, the ellipsoid equation is: （1）； wherein m= r T0 +r R0 / 2, n=TR / 2; r T0 denotes the distance from the transmitter to the scene center point P 0 x 0, y 0,0) along the radar beam center direction, r R0 denotes the distance from the receiver to the point P 0 x 0, y 0,0) with the slant angle θ 0denotes; and based on this, the equation of the intersection ellipse is obtained as:​ （2）； wherein the semi-major axis of the ellipse is a a and the semi-minor axis is b c are derived from equation (1), and the eccentricity of the ellipse is e equal to c / a ; The azimuth space-varying relationship of the received squint angle is derived by solving the geometric relationship in the ellipsoid analytical model: （3）； wherein the linear term coefficient U 1 and V 1 are as follows: （4）； v represents the flight speed of the receiver along the X axis direction; β represents the angle between the projection of the beam on the ground and the azimuth direction; The relationship between the azimuth space-varying received slant ranges is solved as: （5）； where the linear term coefficient Z 1 and the quadratic term coefficient Z 2 are specified as: （6） If take e = 0, h = 0, a = r R0 , , then the received slant range empty variable relationship formula becomes r Rc = r R0 - vsin ( q 0) t c ; the instantaneous range and the expansion coefficients of equation (3) A , B , C , D are all functions of the received slant range r Rc and the received slant angle q , so equation (3) and equation (4) are substituted into the expansion coefficients to re-model, and the Taylor expansion is: （7） wherein (8) In order to realize the residual high-order RCM correction of the azimuth space variation, formula (7) is substituted into the following formula: （25）； wherein f r is the distance frequency; （26） In the above equation, represents the new point target position after Bulk RCMC correction, Δ μ ( t m ; r Rc , t c ) represents the residual higher order RCM error; Finally, formula (9) obtains the expression form of the residual high-order RCM converted into the azimuth space variation component, and the correction method is realized by constructing the following compensation filter: （9） Step S1 is specifically as follows: （10） Multiplying equation (9) by equation (10), we have t m =t c Performing Taylor series expansion at and setting the coefficient of the linear term to zero, we have （11）。 2. The method of claim 1, wherein the method is characterized in that: θ The transmitter is located at ( x T , y T , h The receiver remains stationary at point ) and moves along... X axial direction with velocity v Flight, when direction and time t When = 0, the receiver's position coordinates are (0, 0, ... h The angle between the beam's projection on the ground and its azimuth direction. β For a fixed azimuth angle, r g0 and r g Representing the zero time of azimuth and t c The projected length of the slant distance on the ground is received at all times; the spatial slant angle is... Wherein, It is expressed as follows: （12） According to the geometric relationship, the instantaneous slant range of the target P and is expressed as: （13） Based on this, in t = t c Four Taylor expansions are performed at the point （14） The influence of formula (12) and formula (16) on the simulation parameters is simulated and analyzed, （15） In formula (15), A is a linear distance migration term coefficient, B is a quadratic distance migration term coefficient, C , D is a high-order distance migration term coefficient; Then, the influence of the space-varying squint angle on LRCM and QRCM is quantitatively analyzed by simulation, wherein the error definitions of LRCM and QRCM are as follows: （16） Step S2 is specifically as follows: （17）。 3. The method of claim 2, wherein the method is characterized in that: τ Assuming that the radar transmitter sends a linear frequency modulation signal, then for the point target P ( x , y ,0) the echo signal demodulation form is: （18） wherein, w r (·) is the range envelope, w a (·) is the azimuth envelope, c is the speed of light, f c is the signal carrier frequency, The series inversion is used to convert formula (18) to the range frequency domain: is the range time, K r is the range rate; Wherein, the dominant factor of the range-azimuth coupling is the linear range migration of the first-order term, and the LRCMC is used for processing, and the phase compensation expression is: （19） t=t （20） wherein A 0= -v sin( q 0) denotes a linear distance apodization term coefficient at the center reference point; obtained from the product of equation (20) and equation (19): （21） Resampling by KT transform, i.e. let Step S4 is specifically as follows: m f c / f c +f r , after resampling, Taylor series expansion gives f r ​​ （22） wherein f 0 is an azimuth modulation term, f 1 is a range position term coefficient, f 2 is a range position term, f 3 is a high order range-azimuth coupling term expressed as: （23） Observing equation (22) finds t m with f r The linear coupling term of has been completely eliminated; then, the high-order RCM term is compensated, i.e., the Bulk RCMC and quadratic distance compression processing, and the compensation phase is represented as: （24） wherein B 0, C 0, D 0 denotes the individual distance apodization coefficients at the center reference point, and multiplying equation (22) by equation (24) gives the distance delay profile as: （25）。 4. The method of claim 3, wherein the method is characterized in that: Then, the signal is converted to the azimuth frequency domain and the constant term is ignored: The azimuth modulation term in equation (22) is expanded in a Taylor series about t m =t c and rewritten as: （27） wherein (28) In formula (27), λ = 1 - 2D / c c / f c , f d1 - f d10 is the residual Doppler center frequency, f d2 is the azimuth frequency modulation, f d3 , f d4 represents a high-order term coefficient; the azimuthally-varying relationship derived using the ellipsoidal analytical model is used to re-model each term coefficient, and the expression is as follows:​ （29） The coefficients in formula (30) are expanded into non-azimuth space variation parts and azimuth space variation parts, that is: （30） Next, the high-order non-space variation pre-filtering processing in the frequency domain is introduced, that is, formula (30) is multiplied by the following filter: （31） In order to weaken the azimuth space variation effect, the signal after pre-filtering processing is converted to the azimuth time domain and multiplied by the following time domain phase adjustment factor: （32） ε （33） wherein Y 3、 Y 4 is a pending coefficient; the resulting processing result is Fourier transformed using the MSR to convert the signal into the azimuth frequency domain, multiplied by the FENLCS factor: （34） wherein p 2、 p 3、 p 4 same as the pending coefficient; the multiplied signal is converted to the azimuth time domain to obtain: （35） In formula (35), A Ω is the azimuthal compression term, B Ω contains the focusing position information of the point target, and the other terms are coupling terms that reduce the azimuthal focusing quality; in order to eliminate the influence of the Doppler space-varying parameter, the first-order coupling term B Ω is set to be f d20 π / ε , and the high-order coupling term coefficient is 0, that is: （36） Wherein, in order to make the focusing position and the actual position of the point target always, Substitute formula (36) into formula (35) and multiply by formula (38), that is, the focusing processing in the azimuth direction is completed. The value of is as close to 0.5 as possible; then, solving equation (36) gives: （37） Substituting the above coefficients into A Ω where the final azimuthal compression factor is denoted by （38） It comprises the following modules connected in sequence:

5. An ellipsoid model based near field wide area one station fixed FENLCS imaging system for implementing the ellipsoid model based near field wide area one station fixed FENLCS imaging method of any one of claims 1-4, characterized in that An echo characteristic and squint angle space variation characteristic analysis module: a geometric model of near-range wide-area one-station fixed dual-station SAR imaging is constructed, and the space-varying characteristics of echo characteristics and squint angle are analyzed; A range direction preprocessing module: the target echo signal is preprocessed in the range frequency domain; A range direction processing accuracy improvement module: according to the result of the range direction preprocessing, an ellipsoid model is constructed to obtain the azimuth space-varying relationship of the receiver center slant range and squint angle of the target at any azimuth time point in the same range unit, and the processing accuracy of the range direction is improved by using the residual high-order RCMC method of azimuth space variation; ​ Final focusing image acquisition module: based on the constructed ellipsoid model, the azimuthal space-varying Doppler parameters are remodeled, and the FENLCS algorithm is used to realize the equalization and compression of the azimuthal direction, and the final focusing image is obtained.

6. The near-range wide-area one-station fixed FENLCS imaging system based on an ellipsoidal model as described in claim 5, characterized in that: echo The characteristics and the space-varying characteristics of the squint angle are analyzed as follows: The transmitter is located at ( x T , y T , h The receiver remains stationary at point ) and moves along... X axial direction with velocity v Flight, when direction and time t When = 0, the receiver's position coordinates are (0, 0, ... h ), corresponding r T0 This indicates the distance from the transmitter to the center of the scene along the direction of the radar beam center. P 0( x 0, y The distance from 0,0) r R0 Indicates the receiver's arrival point P 0( x 0, y The distance from 0,0), its oblique angle θ 0 represents; r Tc yes t=t c Transmitter to target P ( x , y, The slope distance of 0). r Rc yes t=t c Time receiver to target P ( x , y, 0) slant range; the angle between the beam projection on the ground and the azimuth direction. β For a fixed azimuth angle, r g0 and r g Representing the zero time of azimuth and t c The projected length of the slant distance on the ground is received at all times; the spatial slant angle is... θ It is expressed as follows: （1） According to the geometric relationship, the instantaneous slant range of the target P and is expressed as: （2） Based on this, in t = t c Four Taylor expansions are performed at the point （3） wherein (4) In formula (4), A is a linear distance migration term coefficient, B is a quadratic distance migration term coefficient, C , D is a high order distance migration term coefficient; According to the simulation parameters, the influence of formula (1) and formula (5) on the generated simulation is analyzed, （5） Subsequently, the influence of the space-varying squint angle on LRCM and QRCM is analyzed quantitatively through simulation, where the error of LRCM and QRCM is defined as follows: （6）。 7. The near field wide area one station fixed FENLCS imaging system based on the ellipsoid model according to claim 6, characterized in that: Distance pre-processing module is as follows: Suppose the radar transmitter sends a linear frequency modulation signal, then for point target P ( x , y ,0) the echo signal demodulation form is: （7） wherein, w r (·) is the range envelope, w a (·) is the azimuth envelope, c is the speed of light, f c is the signal carrier frequency, τ is the range time, K r is the range chirp rate; Convert formula (7) to the distance frequency domain using series inversion: （8） wherein, f r is the distance frequency; the dominant factor of the distance-azimuth coupling is the linear distance migration of the first order term, which is processed by LRCMC, and the phase compensation expression is: （9） wherein A 0= -v sin( q 0) denotes a linear distance apodization term coefficient at the center reference point; obtained from the product of equation (9) and equation (8): （10） Resampling by KT transform, i.e. letting t=t m f c / ( f c +f r ), after resampling, Taylor series expansion gives f r ​ （11） wherein f 0 is an azimuth modulation term, f 1 is a distance position term coefficient, f 2 is a distance azimuth coupling term, and f 3 is a high order distance azimuth coupling term expressed as: （12） Observing equation (11) finds t m with f r The linear coupling term of has been completely eliminated; then, the high-order RCM term is compensated, i.e., the Bulk RCMC and quadratic distance compression processing, and the compensation phase is represented as: （13） wherein B 0, C 0, D 0 denotes the individual distance migration term coefficients at the center reference point, and multiplying equation (13) by equation (11) gives the distance delay curve as: （14） wherein (15) In the above equation, represents the new point target position after Bulk RCMC correction, Δ μ ( t m ; r Rc , t c ) represents the residual higher order RCM error.

8. The near field wide area one station fixed FENLCS imaging system based on the ellipsoid model according to claim 7, characterized in that: Distance processing precision improvement module is as follows: Assume P is a selected imaging point target in the scene area, whose represented distance to the focus position is shifted by Z=r Tc +r Rc The shift is R total (0; r Rc , t c ); wherein, R total (0; r c , t c ) denotes the slant range of the point target at the time of bearing zero from the transmitter, and TP + RP denotes; if the point target P is within the same distance unit as the reference point P 0 after the distance shift, it is considered that TP + RP=TP 0+ RP 0, further constructing a special ellipsoid with the receiver and transmitter positions at the time of zero as two foci, and the half-axes along the Y axis and the Z axis directions are equal, the point target P 0 and P are considered to be on the intersection elliptical surface of the ellipsoid and the XOY plane; Thus, the equation of the ellipsoid is obtained as: （16） wherein, m= r T0 +r R0 / 2, n=TR / 2; and based on this, the equation of the intersecting ellipse is obtained as​ （17） wherein the semi-major axis of the ellipse is a a and the semi-minor axis is b c are given by equation (16), and the eccentricity of the ellipse is e equal to c / a ; The azimuthal space-varying relationship between the received squint distance is solved as: （18） wherein the linear term coefficient U 1 and V 1 are as follows: （19） vsin （20） where the linear term coefficient Z 1 and the quadratic term coefficient Z 2 are specified as: （21） If take e = 0, h = 0, a = r R0 , , then the received slant range empty variable relationship formula becomes r Rc = r R0 - In order to realize the azimuthal space-varying residual high-order RCM correction, formula (23) is substituted into formula (14), and the following is obtained: ( q 0) t c ; the instantaneous range and the expansion coefficients of equation (3) A , B , C , D are all functions of the received range r Rc and the received elevation q , so equation (18) and equation (20) are substituted into the expansion coefficients for re-modeling, and the Taylor expansion is: （22） wherein (23) Formula (24) obtains the expression form of the remaining high-order RCM which has been converted into the azimuthal space-varying component, and its correction method is realized by constructing the following compensation filter: （24） Focusing image acquisition module is as follows: （25） Multiplying equation (24) by equation (25), we have t m =t c Performing Taylor series expansion at x = 0 and setting the coefficient of linear term to zero, we have （26）。 9. The near field wide area one station fixed FENLCS imaging system based on the ellipsoid model according to claim 8, characterized in that the final Subsequently, the signal is converted to the azimuthal frequency domain and the constant term is ignored: The azimuth modulation term in equation (11) is expanded in a Taylor series about t m =t c and rewritten as: （27） wherein (28) In formula (27), λ = 1 - 2D / c c / f c , f d1 - f d10 is a residual Doppler center frequency, f d2 is an azimuth frequency, f d3 , f d4 represents a high-order term coefficient; the azimuthally-varying relationship formula derived using the ellipsoidal analytical model is used to re-model each term coefficient, and the expression is as follows:​ （29） The coefficients in formula (30) are expanded into non-azimuthal space-varying and azimuthal space-varying parts, that is: （30） Next, introduce the frequency domain high-order non-space-varying pre-filtering processing, that is, multiply formula (30) by the following filter: （31） In order to weaken the azimuthal space-varying effect, the signal after pre-filtering processing is converted to the azimuthal time domain and multiplied by the following time domain phase adjustment factor: （32） ε （33） wherein Y 3、 Y 4 is a pending coefficient; the resulting processing result is Fourier transformed using MSR to convert the signal to the azimuth frequency domain, multiplied by the FENLCS factor: （34） wherein p 2、 p 3、 p 4 same as the pending coefficient; convert the multiplied signal to the azimuth time domain to get: （35） In formula (35), A Ω is the azimuthal compression term, B Ω contains the focusing position information of the point target, and the other terms are coupling terms that reduce the azimuthal focusing quality; in order to eliminate the influence of the Doppler space-varying parameter, the first-order coupling term B Ω is set to be f d20 π / ε , and the high-order coupling term coefficient is 0, that is: （36） Wherein, in order to make the focusing position and the actual position of the point target always, Substitute formula (36) into formula (35) and multiply by formula (38), that is, complete the focusing processing in the azimuthal direction. The value of is as close to 0.5 as possible; then, solving equation (36) gives: （37） Substituting the above coefficients into A Ω where the final azimuthal compression factor is given by （38） ​

Images