GNSS-SAR imaging method based on iterative distance vector compression mechanism

Abstract

The application discloses a GNSS-SAR imaging method based on an iterative distance direction compressor mechanism, and comprises the following steps: S1, preprocessing received GNSS-SAR echo signals to generate initial distance direction compressed pulse signals; based on local direct signals, performing autocorrelation operation to generate an iterative distance direction matched filter; S2, using the iterative distance direction matched filter to perform one iteration on the distance direction compressed pulse signals; S3, if the noise of the iteration result is less than a noise threshold value epsilon, then entering S4; if the noise of the iteration result is not less than the noise threshold value epsilon, then entering S2; S4, based on the iteration result with the noise less than the noise threshold value epsilon, performing azimuth direction compression; and S5, splicing the results in each azimuth direction resolution grid to generate a final GNSS-SAR image.

Description

Technical Field

[0001] This invention relates to the field of passive SAR imaging technology, and in particular to a GNSS-SAR imaging method based on an iterative range compression mechanism. Background Technology

[0002] Passive SAR (Synthetic Aperture Radar) systems that utilize GNSS (Global Navigation Satellite System) as an opportunistic transmission source, or GNSS-SAR for short, are significantly cheaper than traditional active SAR systems because they do not require specific signal transmitting devices. They also offer better concealment and resistance to electronic reconnaissance. Furthermore, since GNSS signal transmission is uninterrupted, GNSS-SAR can perform all-weather, blind-spot-free target detection compared to other forms of passive SAR, and has received widespread attention in the past decade.

[0003] GNSS satellites are typically located at very high altitudes. For example, the average distance of a BeiDou satellite is 22,200 km. The GNSS signal strength received at the Earth's surface is extremely weak, typically -130 dBm. For reflected signals used in passive SAR, the signal strength is even weaker, typically -150 dBm to -160 dBm. How to detect targets in such weak signal conditions is a significant and pressing problem that needs to be solved.

[0004] In recent years, researchers have proposed several methods to enhance the signal-to-noise ratio (SNR) of GNSS-SAR images by improving imaging gain. Some researchers have proposed azimuth compression methods based on joint coherent and incoherent integration to improve the azimuth gain of bistatic GNSS-SAR; others have considered multistatic GNSS-SAR models and proposed mechanisms for improving SNR through multi-image incoherent fusion and multi-satellite signal coherent fusion. Although these researchers have proposed some solutions, they cannot effectively address the problem of poor target identification and low image SNR caused by weak reflection signals. Summary of the Invention

[0005] This invention provides a GNSS-SAR imaging method based on an iterative range compression mechanism to solve the technical problem of poor target recognition and low image signal-to-noise ratio caused by weak reflection signals in GNSS-SAR imaging.

[0006] To achieve the above objectives, this invention provides a GNSS-SAR imaging method based on an iterative range compression mechanism, comprising the following steps:

[0007] S1. Signal preprocessing and generation of iterative range-direction matched filter: The received GNSS-SAR echo signal is preprocessed to generate an initial range-direction compressed pulse signal; based on the local direct signal, autocorrelation calculation is performed to generate an iterative range-direction matched filter;

[0008] S2. Perform iteration: Use an iterative range-direction matched filter to iterate the range-direction compressed pulse signal once;

[0009] S3. Filter the iteration results: If the noise in the iteration result is less than the noise threshold ε, proceed to S4; if the noise in the iteration result is not less than the noise threshold ε, proceed to S2.

[0010] S4. Perform azimuth compression: Based on the iteration results where the noise is less than the noise threshold ε, perform azimuth compression.

[0011] S5. Generate GNSS-SAR image: Combine the results in each azimuth resolution grid to generate the final GNSS-SAR image.

[0012] Preferably, in S1, in GNSS-SAR, This represents the k-th echo signal in the range direction time domain when the range is t and the azimuth is u; s m (t,u) represents the local matched filter signal with range t and azimuth u; preprocessing includes: With s m (t,u) performs correlation operations within each code period T:

[0013]

[0014] in, This represents the result of preprocessing. Represents the correlation function of the pseudo-random code; N r This represents the echo signals with different code delays within one code period T. The total number; Indicates echo signal The amplitude value; τ(u) represents a distance in the time domain, s m (t,u) represents the signal propagation delay relative to the GNSS satellite transmitter; Represents the range-direction echo signal in the time domain. Relative to the local matched filter signal s m Propagation delay of (t,u); and They represent echo signals respectively and the local matched filter signal s m The carrier phase of (t,u).

[0015] Preferably, in S1, the iterative distance-oriented matched filter is:

[0016]

[0017] Preferably, in S2, when performing the first iteration, i.e., iterating over the initial distance-direction compression pulse signal, the output result is...

[0018]

[0019] Output the result after the nth iteration. for:

[0020]

[0021]

[0022] Preferably, in S3, when setting the noise threshold ε, the mathematical model is a binary assumption problem:

[0023] H0:T=n r ;

[0024]

[0025] Assume the noise variance is σ v The peak value of pulse compression is μ s The false alarm and detection probabilities corresponding to equation (5) can be expressed as follows:

[0026]

[0027]

[0028] The method of selecting an appropriate threshold value ε can be modeled as an optimization problem:

[0029]

[0030] st 0<ε≤μ s (8)

[0031] The conditions for setting the noise threshold ε can be obtained using an incremental algorithm.

[0032] Preferably, in S4, azimuth compression is performed using a mechanism that combines coherent and incoherent integration:

[0033]

[0034] Among them, T a The duration of integration in the azimuth resolution grid is represented in m. s This represents the duration of the incoherent integral.

[0035] Preferably, in S5, the results from each azimuth resolution grid are combined to generate the final GNSS-SAR image expression as follows:

[0036] I = [I1, I2, ..., I i ,…,I M (10)

[0037] The present invention has the following beneficial effects:

[0038] This invention presents a GNSS-SAR imaging method based on an iterative range compression mechanism. This method effectively preserves carrier phase information even after multiple iterations, avoiding information loss and resulting in good accuracy. By employing a joint coherent and incoherent integration mechanism for azimuth compression, the method optimizes the azimuth signal gain, thereby achieving the best overall imaging gain. This invention enhances the peak value of the compressed pulse signal to maximize the image signal-to-noise ratio, thereby improving the detectability of target objects and extending the detection range of GNSS-SAR.

[0039] In addition to the objectives, features, and advantages described above, the present invention has other objectives, features, and advantages. The invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0040] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0041] Figure 1 This is a flowchart illustrating a preferred embodiment of the present invention.

[0042] Figure 2 This is a simulation model diagram from Experiment 1 of a preferred embodiment of the present invention.

[0043] Figure 3 The result is the latest backscattering algorithm available in Experiment 1 of the preferred embodiment of the present invention.

[0044] Figure 4 The result is the latest backscattering algorithm available in Experiment 1 of the preferred embodiment of the present invention.

[0045] Figure 5 The result is the latest backscattering algorithm available in Experiment 1 of the preferred embodiment of the present invention.

[0046] Figure 6These are the experimental results of the method of the present invention in Experiment 1 of the preferred embodiment of the present invention.

[0047] Figure 7 These are the experimental results of the method of the present invention in Experiment 1 of the preferred embodiment of the present invention.

[0048] Figure 8 These are the experimental results of the method of the present invention in Experiment 1 of the preferred embodiment of the present invention.

[0049] Figure 9 This is a schematic diagram of the experimental scenario for Experiment 2 of the preferred embodiment of the present invention.

[0050] Figure 10 This is the imaging result of the method of the present invention in Experiment 2 of the preferred embodiment of the present invention.

[0051] Figure 11 The imaging results are from Experiment 2 of the preferred embodiment of the present invention, using the latest existing backscattering algorithm.

[0052] In the attached diagram:

[0053] 1. BeiDou B3I satellite; 2. BeiDou receiver; 3. Direct signal receiving antenna; 4. Reflected signal receiving antenna; 5. Ground scattering area; 6. Target object; 7. Direct signal; 8. Reflected signal; 9. Trajectory of synthetic aperture; 10. Antenna support; 11. BeiDou signal intermediate frequency acquisition device; 12. Computer control terminal; 13. Ground surface. Detailed Implementation

[0054] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0055] See Figure 1 In a preferred embodiment of the present invention, a GNSS-SAR imaging method based on an iterative range compression mechanism is provided, comprising the following steps:

[0056] S1. Signal preprocessing and generation of iterative range-direction matched filter: The received GNSS-SAR echo signal is preprocessed to generate an initial range-direction compressed pulse signal; based on the local direct signal, autocorrelation calculation is performed to generate an iterative range-direction matched filter.

[0057] In S1, in GNSS-SAR, This represents the k-th echo signal in the range direction time domain when the range is t and the azimuth is u; s m (t,u) represents the local matched filter signal with range t and azimuth u; preprocessing includes: With s m (t,u) performs correlation operations within each code period T:

[0058]

[0059]

[0060] in, This represents the result of preprocessing. Represents the correlation function of the pseudo-random code; N r This represents the echo signals with different code delays within one code period T. The total number; Indicates echo signal The amplitude value; τ(u) represents a distance in the time domain, s m (t,u) represents the signal propagation delay relative to the GNSS satellite transmitter; Represents the range-direction echo signal in the time domain. Relative to the local matched filter signal s m Propagation delay of (t,u); and They represent echo signals respectively and the local matched filter signal s m The carrier phase of (t,u) in the same distance direction in the time domain and It is usually constant. As can be seen from equation (1), the range-direction compressed signal is mainly composed of a set of orthogonal phases.

[0061] In S1, the iterative distance-oriented matched filter is:

[0062]

[0063] S2. Perform iteration: Use an iterative range-direction matched filter to perform one iteration on the range-direction compressed pulse signal.

[0064] In S2, during the first iteration, which iterates over the initial range compression pulse signal, the output result is...

[0065]

[0066] Output the result after the nth iteration. for:

[0067]

[0068] As can be seen from equation (4), after n iterations, the carrier phase information The information has been well preserved, avoiding data loss and ensuring the accuracy of this solution.

[0069] S3. Filtering the iteration results: If the noise in the iteration result is less than the noise threshold ε, proceed to S4; if the noise in the iteration result is not less than the noise threshold ε, proceed to S3.

[0070] In S3, when setting the noise threshold ε, the mathematical model is a problem with two assumptions:

[0071] H0:T=n r ;

[0072]

[0073] Assume the noise variance is σ v The peak value of pulse compression is μ s Since the number of samples in the time domain is very large, the distribution of equation (5) can be approximated as a Gaussian distribution. Therefore, the false alarm and detection probabilities corresponding to equation (5) can be expressed as follows:

[0074]

[0075]

[0076] The method of selecting an appropriate threshold value ε can be modeled as an optimization problem:

[0077]

[0078] st 0<ε≤μ s (8)

[0079] By analyzing the properties of equation (8), it is found that equation (8) is a non-convex optimization problem, and a global optimal solution cannot be obtained. Therefore, it is necessary to determine the specific σ in the current environment. v The value of ε is used to find a local optimum. The incremental algorithm, a common algorithm for solving non-convex optimization problems, can be used to determine the conditions for setting the noise threshold ε.

[0080] S4. Perform azimuth compression: Based on the iteration results where the noise is less than the noise threshold ε, perform azimuth compression.

[0081] In S4, azimuth compression is performed using a mechanism that combines coherent and incoherent integration:

[0082]

[0083] Among them, T a The duration of integration in the azimuth resolution grid is represented in m. s This represents the duration of the incoherent integral.

[0084] By employing a mechanism that combines coherent and incoherent integration for azimuth compression, this scheme can optimize the signal gain in the azimuth direction, thereby achieving the best overall imaging gain.

[0085] S5. Generate GNSS-SAR image: Combine the results in each azimuth resolution grid to generate the final GNSS-SAR image.

[0086] In S5, the results from each azimuth resolution grid are combined to generate the final GNSS-SAR image expression as follows:

[0087] I = [I1, I2, ..., I i ,…,I M (10)

[0088] In a preferred embodiment of the present invention, the method proposed in this invention is verified through simulation and experimental data testing using BeiDou B3I signals as an opportunistic transmission source, i.e., Experiment 1. The parameter settings for this experiment are shown in Table 1. The sampling rate in the range direction is 40MHz, the period T of the BeiDou B3I C / A code signal is 1ms, the code frequency is 10.23MHz, and the propagation speed is c = 3 × 10⁻⁶. 8 m / s. The simulation model used in this experiment is referred to [reference needed]. Figure 2 In the simulation model, BeiDou B3I satellite 1 acts as the signal source, transmitting signals to the remote sensing area. At the port of BeiDou receiver 2, direct signal receiving antenna 3 receives the direct signal 7 from BeiDou B3I satellite 1 for synchronization, and reflected signal receiving antenna 4 receives the reflected signal 8 from target object 6 for passive radar imaging. Target object 6 consists of two strong ground reflection areas located in ground scattering area 5; the distance between target object 6 and BeiDou receiver 2 is 3m. The trajectory 9 of the synthetic aperture is generated based on the curved motion of reflected signal receiving antenna 4. The synthetic aperture is mounted on BeiDou receiver 2, and its trajectory length is 100°.

[0089] In a preferred embodiment of the present invention, the results of the latest back projection (BP) algorithm can be found in [reference needed]. Figures 3 to 5 Based on the parameters in Table 1 and Figure 2 Simulation experiments were conducted using the simulation model at signal-to-noise ratios of -45dB, -50dB, and -55dB. The experimental results are shown in [reference needed]. Figures 6 to 8 By comparing the results with the current state-of-the-art backscattering algorithms ( Figures 3 to 5 Comparing pixel density, specifically Figure 3 and Figure 6 Compare Figure 4 and Figure 7 Compare and Figure 5 and Figure 8A comparison reveals that the method of this invention significantly improves pixel density, meaning it provides a significantly higher imaging signal-to-noise ratio. Specifically, even when the signal-to-noise ratio drops to -55dB, the method of this invention can still clearly detect two target objects, whereas in the latest BP imaging method, two strongly reflective objects are difficult to identify. This demonstrates that the method proposed in this invention has a higher degree of target detection capability.

[0090] Table 1: Experimental Parameters

[0091] signal frequency 1268.52MHz Signal bandwidth B 10.23MHz Pseudo-random code period 1ms signal propagation speed <![CDATA[3×10 8 m / s]]> Distance sampling rate 40MHz Platform rotation speed 1° / s Experimental ambient temperature 300K

[0092] In a preferred embodiment of the present invention, to further verify the method of the present invention, an experiment based on measured satellite signal data from BeiDou B3I was also conducted, namely Experiment 2. See the experimental scenario below. Figure 9 The BeiDou B3I satellite 1 serves as the signal source, transmitting signals to the remote sensing area. A direct signal receiving antenna 3 and a reflected signal receiving antenna 4 are mounted on an antenna support 10, which is located on the ground surface 13. The direct signal receiving antenna 3 receives the direct signal 7 for synchronization; the reflected signal receiving antenna 4 receives the reflected signal 8 from the remote sensing area for radar imaging. In this experiment, a computer control terminal 12 controls the BeiDou signal intermediate frequency acquisition unit 11 to acquire the signals received by the antennas. The target object 6 consists of two identical reflectors, each with a cross-sectional area of ​​approximately 0.2 m². 2 Made of tin foil (44cm×44cm), with the location of antenna bracket 10 as the distance 0 point, the horizontal distance between the two reflectors and antenna bracket 10 is 3m; see imaging results. Figure 10 , where the horizontal axis represents the distance in the range direction and the vertical axis represents the azimuth angular distance.

[0093] In a preferred embodiment of the present invention, the imaging results of the latest backscattering algorithm are shown below. Figure 11 By using the imaging results of this method ( Figure 10 Imaging results compared with the latest backscattering algorithms ( Figure 11 Based on the comparison of pixel density, it can be seen that the method of the present invention significantly improves the imaging gain, which means that the method of the present invention is more sensitive to weak signals than the latest existing BP imaging method and can provide higher target detectability.

[0094] In summary, the preferred embodiment of the GNSS-SAR imaging method based on an iterative range compression mechanism of this invention proposes an imaging method based on an iterative range compression mechanism. In this scheme, even after multiple iterations, the carrier phase information can still be well preserved, avoiding information loss and resulting in good accuracy. The use of a joint coherent and incoherent integration mechanism for azimuth compression optimizes the azimuth signal gain, thereby achieving the best overall imaging gain. This invention can enhance the peak value of the compressed pulse signal to maximize the image signal-to-noise ratio, thereby improving the detectability of target objects and extending the detection range of GNSS-SAR.

[0095] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A GNSS-SAR imaging method based on an iterative range compression mechanism, characterized in that, Includes the following steps: S1. Signal preprocessing and generation of iterative range-guided matched filter: The received GNSS-SAR echo signal is preprocessed to generate an initial range-guided compressed pulse signal; Based on the local direct signal, autocorrelation calculation is performed to generate an iterative range-oriented matched filter; S2. Perform iteration: Use the iterative range-direction matched filter to perform one iteration on the range-direction compressed pulse signal; S3. Filtering iteration results: If the noise in the iteration result is less than the noise threshold value. ε Then proceed to S4; If the noise in the iteration result is not less than the noise threshold value ε Then proceed to S2; S4. Perform azimuth compression: based on noise being less than the noise threshold. ε The iterative results are then subjected to azimuth compression. S5. Generate GNSS-SAR image: Combine the results in each azimuth resolution grid to generate the final GNSS-SAR image; In S1, in GNSS-SAR, Indicates distance direction Orientation At that time, in the distance to the time domain, the first One echo signal; Indicates distance direction Orientation The local matched filter signal at that time; the preprocessing includes: and In each code cycle Perform related calculations within: d （1） in, This represents the result of preprocessing. Represents the function related to pseudo-random codes; Indicates one code period Echo signals with different code delays The total number; Indicates echo signal The amplitude value; Represents a distance in the time domain, The signal propagation delay relative to the GNSS satellite transmitter; Represents the range-direction echo signal in the time domain. relative to the local matched filter signal The propagation delay; and They represent echo signals respectively Locally matched filter signal The carrier phase.

2. The GNSS-SAR imaging method based on iterative range compression mechanism according to claim 1, characterized in that, In S1, the iterative distance-oriented matched filter is: （2）。 3. The GNSS-SAR imaging method based on iterative range compression mechanism according to claim 2, characterized in that, In S2, during the first iteration, i.e., when iterating over the initial distance-direction compression pulse signal, the output result is... : （3） Output the result after the nth iteration. for: （4）。 4. The GNSS-SAR imaging method based on iterative range compression mechanism according to claim 3, characterized in that, In S3, the noise threshold value is set. ε In mathematics, this is modeled as a problem with two assumptions: : ; : （5） Assume the noise variance is The peak value of pulse compression is The false alarm and detection probabilities corresponding to equation (5) can be expressed as follows: （6） （7） Select an appropriate threshold value The method can be modeled as an optimization problem: （8） The set noise threshold value can be obtained using an incremental algorithm. ε conditions.

5. The GNSS-SAR imaging method based on iterative range compression mechanism according to claim 4, characterized in that, In S4, azimuth compression is performed using a mechanism that combines coherent and incoherent integration: （9） in, This represents the integration time within the azimuth resolution grid. This represents the duration of the incoherent integral.

6. The GNSS-SAR imaging method based on iterative range compression mechanism according to claim 5, characterized in that, In S5, the results from each azimuth resolution grid are combined to generate the final GNSS-SAR image expression as follows: （10）。

Images