A frequency-modulated stepped radar range image extraction method based on same-distance redundant echo amplitude-phase screening

Abstract

The application discloses a frequency-modulated stepped radar range image extraction method based on same-distance redundant echo amplitude-phase screening and relates to the technical field of radar signal processing. First, complex distance data formed by inverse fast Fourier transform of frequency-modulated stepped radar echo is acquired, and data corresponding to the same physical distance unit is composed into a same-distance candidate echo group. Deterministic distance phase items are determined according to coarse distance sampling positions, and distance phase compensation is carried out. Residual phase differences between candidate echoes are calculated, and phase-consistent candidate echoes are determined according to a phase difference threshold. Extraction echoes are determined according to amplitudes, and one-dimensional high-resolution range images are formed in the order of physical distances. The application increases phase relation discrimination and reduces the possibility that abnormal amplitude candidates are directly used as extraction echoes.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, and in particular to a method for range image extraction of frequency-modulated stepping radar based on the screening of amplitude and phase of redundant echoes at the same distance. Background Technology

[0002] Frequency-modulated stepping radar uses linear frequency-modulated subpulses with a carrier frequency varying at predetermined step intervals to form the transmitted signal, and obtains the synthetic bandwidth by stepping the frequency of multiple subpulses. After pulse compression and motion compensation, the received echo can be subjected to inverse fast Fourier transform along the step frequency dimension to obtain a one-dimensional high-resolution range image of the target. Since there is redundant range information that needs to be removed in the high-resolution processing result, target extraction processing is also required during the range image formation process. The article "Frequency-Stepping Radar Parameter Design and Target Extraction Algorithm" published in "Systems Engineering and Electronics Technology" (Vol. 23, No. 6, pp. 26-31) studies the target extraction criteria and processing methods of frequency-stepping radar and analyzes the impact of target motion on target extraction processing.

[0003] To address redundant range information in high-resolution processing results, existing technologies employ methods such as selecting the largest value within the same range to extract the range image. Chinese invention patent application CN115902885A discloses a step-frequency high-resolution short-range radar detection system and method based on FPGA. This system performs an inverse fast Fourier transform (IFT) on the echo data within the range gate to obtain one-dimensional range image amplitudes, and extracts valid data from each group of IFT results. After aligning the valid data according to range positions, the amplitudes of data corresponding to the same range position are compared, and the data with the larger amplitude is retained to complete the selection of the largest value within the same range for redundancy removal. The redundancy removal results from different range gates are then rearranged and stitched together to form a one-dimensional high-resolution range image.

[0004] Existing methods for selecting the largest possible distance echo use the amplitude of candidate data corresponding to the same distance location as the basis for range image extraction. When there are multiple redundant echoes that can participate in the selection for the same physical distance cell, the extracted echo for the corresponding physical distance cell is still determined according to the amplitude, lacking a means of candidate echo discrimination other than amplitude comparison. Summary of the Invention

[0005] The purpose of this invention is to provide a range image extraction method for frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance. This method can solve the problem that existing methods for selecting echoes at the same distance only determine the extracted echo based on amplitude when there are multiple candidate echoes in the same physical range cell, and lack other means of candidate discrimination besides amplitude. This method can achieve range image extraction based on amplitude after screening candidate echoes based on the remaining phase difference.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for range image extraction of frequency-modulated stepping radar based on range-redundant echo amplitude and phase filtering includes: Multiple sets of complex range data are obtained from the frequency-modulated step radar echo after inverse fast Fourier transform. Each set of complex range data corresponds to a coarse range sampling position. Based on the physical distance corresponding to each group of complex distance data, the complex data corresponding to the same physical distance unit are determined as candidate echoes of the same distance, forming a candidate echo group of the same distance; The deterministic range phase term of each candidate echo at the same distance is determined based on the distance offset between the coarse range sampling position corresponding to each candidate echo at the same distance and the physical distance of the corresponding physical range unit. Based on the deterministic range-phase term, range-phase compensation is performed on the candidate echo group containing multiple candidate echoes at the same range to obtain candidate echoes with unified phase reference. Calculate the remaining phase difference between candidate echoes with the same phase reference and corresponding to the same distance candidate echo group; Each remaining phase difference is compared with the phase difference threshold, and the phase-consistent candidate echoes corresponding to the candidate echo group at the same distance are determined based on the comparison results. When a phase-consistent candidate echo exists, the extracted echo of the corresponding physical distance cell is determined based on the amplitude of the phase-consistent candidate echo. When no phase-consistent candidate echo is formed, the extracted echo of the corresponding physical distance unit is determined based on the amplitude of each candidate echo at the same distance within the corresponding candidate echo group at the same distance. Echoes are extracted according to the distance order of each physical distance unit to obtain a one-dimensional high-resolution range image.

[0007] Preferably, multiple sets of complex range data are acquired by obtaining the frequency-modulated stepping radar echo through inverse fast Fourier transform, with each set of complex range data corresponding to a coarse range sampling position, including: The frequency-modulated step radar echoes are grouped according to the coarse range sampling positions to obtain the step frequency echo sequences corresponding to each coarse range sampling position. Inverse Fast Fourier Transform (IFFT) is performed on each step frequency echo sequence along the step frequency dimension. The real and imaginary parts of the IFFT result are retained, and the real and imaginary parts corresponding to the same coarse distance sampling position are combined into a set of complex distance data.

[0008] Preferably, based on the physical distance corresponding to each group of complex distance data, the complex data corresponding to the same physical distance unit are determined as candidate echoes of the same distance, forming a candidate echo group of the same distance, including: Based on the coarse distance sampling position corresponding to each group of complex distance data, the distance sequence number of each complex data in the group, and the distance cell interval after inverse fast Fourier transform, determine the physical distance corresponding to each complex data. The complex data corresponding to the same physical distance cell are distance aligned, and the distance-aligned complex data are identified as candidate echoes of the same distance. Candidate echoes of the same distance corresponding to the same physical distance unit are grouped into a candidate echo group of the same distance.

[0009] Preferably, the deterministic range phase term of each candidate echo at the same distance is determined based on the distance offset between the coarse range sampling position corresponding to each candidate echo at the same distance and the physical distance of the corresponding physical range cell, including: Using the physical distance of the corresponding physical distance unit as a unified distance benchmark, the distance offset of the coarse distance sampling position corresponding to each candidate echo at the same distance relative to the unified distance benchmark is determined; Based on the range offset and the range-phase relationship of the frequency-modulated step radar echo, the deterministic range-phase term corresponding to each candidate echo at the same range is determined.

[0010] Preferably, based on a deterministic range-phase term, range-phase compensation is performed on a group of candidate echoes containing multiple candidate echoes at the same range to obtain candidate echoes with a unified phase reference, including: Based on the deterministic range phase term corresponding to each candidate echo at the same distance, construct complex phase compensation factors with a modulus of 1 and a phase angle that is the opposite of the corresponding deterministic range phase term. Each candidate echo at the same distance is multiplied by its corresponding complex phase compensation factor so that each candidate echo in the same distance candidate echo group corresponds to a unified distance reference, thus obtaining candidate echoes with a unified phase reference.

[0011] Preferably, calculating the remaining phase difference between candidate echoes with a unified phase reference corresponding to the same candidate echo group at the same distance includes: For any two candidate echoes with the same phase reference within the same candidate echo group, calculate the complex conjugate product of one candidate echo and the other candidate echo; The absolute value of the principal argument of the complex conjugate product is taken as the residual phase difference between the two candidate echoes with unified phase references. The residual phase difference satisfies: In the formula, and They represent the first The first physical distance unit corresponding to the first The and the first Candidate echoes with unified phase reference, and For different candidate echo numbers, express The complex conjugate, This indicates taking the principal argument. This represents the remaining phase difference between two candidate echoes with a unified phase reference.

[0012] Preferably, each remaining phase difference is compared with a phase difference threshold, and the phase-consistent candidate echoes corresponding to the candidate echo group at the same distance are determined based on the comparison results, including: For each candidate echo with a unified phase reference, the remaining phase difference and phase difference threshold between the candidate echo with the unified phase reference and the other candidate echoes with the unified phase reference in the same distance candidate echo group are compared. When there is at least one remaining phase difference that is not greater than the phase difference threshold, the candidate echo with the corresponding unified phase reference is retained; when all remaining phase differences are greater than the phase difference threshold, the candidate echo with the corresponding unified phase reference is filtered out. Candidate echoes with consistent phase references are identified as phase-consistent candidate echoes.

[0013] Preferably, each remaining phase difference is compared with a phase difference threshold, and the phase-consistent candidate echoes corresponding to the candidate echo group at the same distance are determined based on the comparison results, including: When there are more than three candidate echoes with the same phase reference in the same distance candidate echo group, count the number of remaining phase differences between each candidate echo with the same phase reference and the other candidate echoes with the same phase reference that are not greater than the phase difference threshold. Determine the maximum value among the remaining phase differences corresponding to the candidate echoes with unified phase references; When the maximum value is greater than zero, the candidate echo with the largest amplitude is selected from the candidate echoes with the same number of remaining phase differences as the maximum value as the reference candidate echo. The candidate echo that is consistent with the reference candidate echo and the candidate echo whose remaining phase difference with the reference candidate echo is not greater than the phase difference threshold is defined as the phase-consistent candidate echo. When the maximum value is zero, it is determined that no candidate echo with phase consistency has been formed in the candidate echo group at the same distance.

[0014] Preferably, when phase-consistent candidate echoes exist, determining the extracted echo of the corresponding physical distance cell based on the amplitude of the phase-consistent candidate echoes includes: The corresponding complex modulus is determined based on the real and imaginary data of each phase-consistent candidate echo, and the complex modulus is used as the amplitude of the corresponding phase-consistent candidate echo. Compare the amplitudes of each phase-consistent candidate echo corresponding to the same physical distance cell, and determine the phase-consistent candidate echo with the largest amplitude as the extracted echo of the corresponding physical distance cell.

[0015] Preferably, when no phase-consistent candidate echo is formed, the extracted echo of the corresponding physical distance unit is determined based on the amplitude of each candidate echo within the corresponding candidate echo group, including: When the candidate echo group at the same distance contains only one candidate echo at the same distance, the candidate echo at the same distance is determined as the extracted echo of the corresponding physical distance unit. When a candidate echo group with the same distance contains multiple candidate echoes with the same distance but does not form a candidate echo with the same phase, the corresponding complex modulus value is determined based on the real part data and imaginary part data of each candidate echo with the same distance, and the candidate echo with the largest complex modulus value with the same distance is determined as the extracted echo of the corresponding physical distance unit.

[0016] The present invention discloses the following beneficial effects: (1) This invention determines complex data corresponding to the same physical distance unit as candidate echoes of the same distance, and determines a deterministic distance phase term based on the distance offset between the coarse distance sampling position and the physical distance. Distance phase compensation is then performed within the candidate echo group of the same distance, so that candidate echoes from different coarse distance sampling positions correspond to a unified phase reference. Distance phase compensation reduces the influence of the deterministic phase term formed by the difference in the distance reference on the phase comparison of candidate echoes, allowing the remaining phase difference to be used to characterize the relative phase relationship between candidate echoes of the same distance.

[0017] (2) This invention calculates the remaining phase difference between candidate echoes with a unified phase reference, compares the remaining phase difference with a phase difference threshold, determines phase-consistent candidate echoes from the same distance candidate echo group, and then determines the extracted echo based on the amplitude of the phase-consistent candidate echoes. Compared with the same-distance selection method that only determines the extracted echo based on amplitude, this invention adds phase relationship discrimination when multiple candidate echoes correspond to the same physical distance unit. When a same-distance candidate echo group contains K candidate echoes, pairwise comparisons will form at most K(K-1) / 2 remaining phase differences, so that the determination of candidate echoes no longer depends solely on the magnitude of a single amplitude, reducing the possibility of abnormal amplitude candidates being directly used as extracted echoes.

[0018] (3) In this invention, when phase-consistent candidate echoes are formed, the extracted echoes are determined based on the amplitude of the phase-consistent candidate echoes. When phase-consistent candidate echoes are not formed, the extracted echoes are determined based on the amplitude of each candidate echo within the same distance candidate echo group. Both processing results correspond to the determined physical distance units, avoiding the lack of extracted echoes for the corresponding physical distance units when phase screening fails to form candidate results. The extracted echoes are arranged in the distance order of the physical distance units to form a one-dimensional high-resolution distance image. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart of the method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of candidate echo distance alignment provided in an embodiment of the present invention; Figure 3 This is a comparison diagram of the remaining phase difference provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the system structure provided in an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Figure 1 The method flowchart provided in the embodiments of the present invention is as follows: Figure 1 As shown, this embodiment provides a method for range image extraction of frequency-modulated stepping radar based on range-redundant echo amplitude and phase filtering, including: Step 100: Acquire multiple sets of complex range data formed by inverse fast Fourier transform of the frequency modulation stepping radar echo, with each set of complex range data corresponding to a coarse range sampling position; Step 101: Based on the physical distance corresponding to each group of complex distance data, determine the complex data corresponding to the same physical distance unit as candidate echoes of the same distance, and form a candidate echo group of the same distance; Step 102: Determine the deterministic range phase term of each candidate echo at the same distance based on the distance offset between the coarse range sampling position corresponding to each candidate echo at the same distance and the physical distance of the corresponding physical range cell; Step 103: Based on the deterministic range-phase term, perform range-phase compensation on the candidate echo group containing multiple candidate echoes of the same range to obtain candidate echoes with unified phase reference; Step 104: Calculate the remaining phase difference between candidate echoes with the same phase reference for each candidate echo group at the same distance; Step 105: Compare each remaining phase difference with the phase difference threshold, and determine the phase-consistent candidate echoes corresponding to the candidate echo groups at the same distance based on the comparison results; Step 106: When there is a phase-consistent candidate echo, determine the extracted echo of the corresponding physical distance unit based on the amplitude of the phase-consistent candidate echo; Step 107: When no phase-consistent candidate echo is formed, determine the extracted echo of the corresponding physical distance unit based on the amplitude of each candidate echo within the corresponding candidate echo group. Step 108: Arrange the echoes according to the distance order of each physical distance unit to obtain a one-dimensional high-resolution range image.

[0024] In one specific implementation, the frequency-modulated stepping radar sequentially generates multiple step frequencies at predetermined frequency intervals within one coherent processing cycle. The echoes after orthogonal demodulation are retained as complex baseband data. When the range change of the target within one coherent processing cycle is not negligible, motion phase compensation is performed according to the already obtained target motion parameters before proceeding to step 100; when the range change of the target within one coherent processing cycle is negligible, the coherent baseband echoes are directly used for subsequent processing. Each set of echoes entering step 100 uses the same step frequency sequence and maintains the coherence relationship of the step frequency dimension.

[0025] Let a coherent processing cycle contain The number of equally spaced step frequencies, the first Each step frequency is represented as: In the formula, Indicates the first One step frequency, Indicates the first step frequency, This indicates the frequency interval between adjacent step frequencies. Indicates the step frequency sequence number. This indicates the number of step frequencies involved in a single high-resolution distance processing operation.

[0026] Complex baseband echoes are grouped according to their coarse-range sampling locations. The same coarse-range sampling location is... The complex echoes obtained at each step frequency are arranged according to the step frequency number to form the corresponding step frequency echo sequence. The step frequency echo sequence corresponding to each coarse-distance sampling position is denoted as , Indicates the coarse distance sampling position number.

[0027] In one data formation method, when forming a stepped frequency echo sequence according to coarse-range sampling positions, the corresponding coarse-range sampling positions are used as local distance references. Let the first... The physical distance corresponding to each coarse-distance sampling location is The corresponding two-way reference delay is: In the formula, Indicates the first Two-way reference delay corresponding to each coarse distance sampling position This indicates the speed of electromagnetic wave propagation.

[0028] Local range references can be formed using the corresponding phase reference during range gate data reading, or equivalent complex transformations can be performed on complex baseband echoes. The local distance reference conversion factor corresponding to each step frequency is denoted as . ,satisfy: Let the complex baseband echo before the local range reference transformation be denoted as . The converted step frequency echo is: When using a single-station radar round-trip propagation phase convention, the physical distance is The range-phase term corresponding to the echo is expressed as: After local range reference transformation, the range phase term becomes: Therefore, the range phase in the same set of step frequency echoes corresponds to the coarse range sampling position. As a local distance reference. The modulus is 1, and the local distance reference conversion does not change the amplitude of the complex baseband echo.

[0029] Along the step frequency dimension conduct Point-inverse fast Fourier transform, the first The transformation result is represented as follows: In the formula, , Indicates the first The first coarse-distance sampling position corresponding to the first A complex distance data point, It represents the imaginary unit.

[0030] When the inverse fast Fourier transform uses other normalization methods with a uniform ratio, each complex distance data will simultaneously produce the same amplitude ratio change, without changing the relative phase between candidate echoes or the subsequent amplitude ranking relationship.

[0031] Unlike distance image extraction methods that only save the amplitude of the inverse fast Fourier transform result, this embodiment retains... Real and imaginary data: In the formula, Represents the real part of the data. Represents the imaginary part of the data. This represents the imaginary unit. The real and imaginary data together form the data basis for subsequent range-phase compensation and residual phase difference calculations.

[0032] In this embodiment, the parameters listed in Table 1 are used to calculate the distance coordinates.

[0033] Table 1 Range Processing Parameters of Frequency Modulated Stepping Radar When 32 equally spaced step frequencies are used with a 32-point inverse fast Fourier transform, the corresponding high-resolution distance grid spacing is: Substituting the parameters into Table 1, we get: Here This indicates the distance grid spacing corresponding to the inverse fast Fourier transform. When zero-padding is used to increase the number of points in the inverse fast Fourier transform, the distance grid becomes finer as the number of transform points increases, but the actual target distance resolution is still limited by the effective synthesis bandwidth.

[0034] The frequency step interval also determines the single-point unambiguous distance in the step frequency dimension: Table 1 Parameter Correspondence: And satisfy: Complex baseband sampling rate When the frequency is 96MHz, the distance sampling interval between adjacent coarse distance sampling positions is: Sub-pulse bandwidth For 24MHz, the corresponding coarse range resolution scale is: The distance oversampling factor in this embodiment is: Within a coarse distance resolution scale, there are four distance sampling locations. The distance between adjacent coarse distance sampling locations is 1.5625m, which is less than 6.25m. Therefore, the high-resolution distance data formed by adjacent coarse distance sampling locations overlap. Consequently, the same physical distance unit may acquire multiple complex distance data from different coarse distance sampling locations.

[0035] Step 101: Establish a unified physical distance coordinate system.

[0036] Let the first The physical distance corresponding to each coarse-distance sampling location is After performing distance folding and rearranging on the inverse fast Fourier transform results, the following is used: Indicates and The corresponding distance index, then the first The distance offset of each complex distance data point relative to the coarse distance sampling position is expressed as: No. Group 1 The physical distances corresponding to each complex distance data point are: The inverse fast Fourier transform results have the following characteristics: The distance loop is periodic. When valid distance data crosses an unambiguous distance boundary, the data arrangement order is first adjusted according to the direction of continuous increase in physical distance, and then determined. Data after the fold position is still arranged according to the actual physical distance, and does not directly participate in the same distance grouping according to the original sequence number after the inverse fast Fourier transform array wraps around.

[0037] Let the initial distance of the one-dimensional high-resolution range image to be formed be . , No. The center distance of each physical distance unit is: In the formula, Indicates the physical distance unit number.

[0038] when satisfy: The corresponding complex distance data Classified as Each physical distance unit. The left-closed, right-open distance interval ensures that data located at the boundary between adjacent physical distance units enters only one physical distance unit.

[0039] When complex distance data from different coarse distance sampling locations fall into the same physical distance cell after the above distance mapping, they are respectively identified as candidate echoes of the same distance. There are a total of physical distance units. When there are multiple candidate echoes at the same distance, it is denoted as: The grouping of candidate echoes at the same distance is based on the same physical distance coordinates, rather than the same array index in the inverse fast Fourier transform results of different groups.

[0040] Figure 2 This is a schematic diagram illustrating the distance alignment of candidate echoes at the same distance. (Example:) Figure 2 As shown, the four consecutive rows of distance cells correspond to the coarse distance sampling positions. , , and The row distance units are arranged from left to right along the direction of increasing physical distance; the vertical dashed lines correspond to the physical distance units. The position, and passing through the distance cells aligned with the physical distance cells in the four rows, the corresponding complex distance data are sequentially marked as... , , and The four marked positions are at the same physical distance position, which are used to represent the correspondence between complex distance data from different coarse distance sampling positions and the same physical distance unit. The arrow at the bottom of the figure indicates that the physical distance increases from left to right.

[0041] In one distance alignment method, distance alignment is performed in the complex domain without first converting the complex distance data into amplitude-only data. When the physical distance corresponding to the complex distance data does not completely coincide with the center of the physical distance cell, the same interpolation coefficient is used for the real and imaginary parts of adjacent complex distance data. Let the physical distance cell... Data located in the same set of adjacent complex distances and Corresponding physical distance and Between them, the interpolation coefficients are: In the formula, Represents physical distance unit The corresponding complex distance interpolation coefficients. The distance-aligned complex candidate values ​​are represented as: The real and imaginary parts use the same Interpolation is performed, but no additional amplitude or phase interpolation weights are set during the distance alignment process. When the physical distance coincides with the distance grid, the corresponding complex distance data is used directly without interpolation.

[0042] After completing the distance alignment, it is also necessary to address the phase reference differences caused by different coarse distance sampling positions.

[0043] In this embodiment, each set of complex distance data uses the corresponding coarse distance sampling position as a local distance reference. Let the first... The candidate echoes at the same distance originate from the coarse-distance sampling location. , No. The distance offset between each physical distance cell and its corresponding coarse distance sampling position is: The round-trip propagation range of a single-station radar is twice the target range. When using a complex representation where the baseband phase decreases with the increase in echo range, the... At each step frequency, the range phase term corresponding to the range offset is expressed as: Separating the phase term corresponding to the first step frequency from the phase term that changes with the step frequency index, we obtain: The second exponential term follows the step frequency index. The phase change is linear, corresponding to the distance position after the inverse fast Fourier transform. Step 101 has already completed the distance alignment of different groups of complex distance data according to the physical distance. The first exponential term is related to the local distance reference used in each group, and exhibits a deterministic phase shift between candidate echoes of the same distance formed at different coarse distance sampling positions.

[0044] The second exponential term mentioned above varies with the step frequency index, determining the distance position of the complex distance data after the inverse fast Fourier transform. After step 101 uses unified physical distance coordinates and completes distance alignment in the complex domain, step 102 no longer compensates for the linear phase term, but determines the deterministic distance phase term based on the constant distance phase term formed by the first step frequency and different local distance references.

[0045] Step 102 will be the first The deterministic range-phase term corresponding to each candidate echo at the same distance is denoted as: In the formula, , , and All are known quantities. The deterministic range-phase term is calculated directly from the range position and radar operating frequency, without depending on the candidate echo amplitude.

[0046] When the carrier frequency is high or the distance offset is large It may contain multiple complete phase cycles. During numerical calculations, the deterministic distance phase term is transformed to the principal value interval: Therefore, we can conclude that: Step 103 According to Construct a complex phase compensation factor: Expand into real and imaginary parts: No. After range-phase compensation, the candidate echoes at the same distance are represented as follows: In the formula, This indicates a candidate echo with a unified phase reference.

[0047] The magnitude of the complex phase compensation factor satisfies: Therefore: The magnitude values ​​of candidate echoes remain unchanged before and after distance and phase compensation. Unifying the phase reference does not change the original amplitude relationship between candidate echoes at the same distance, and subsequent amplitude comparisons are still based on the amplitude of the echoes themselves.

[0048] In receivers employing an inverse quadrature demodulation symbol convention, the signs of the range phase term and the compensation phase change simultaneously. As long as the range phase term and the complex phase compensation factor use the same complex signal convention, the remaining phase difference corresponding to the subsequent complex conjugate product is unaffected by the symbol convention.

[0049] Step 104: Calculate the remaining phase difference between candidate echoes with a unified phase reference within the same candidate echo group at the same distance.

[0050] Suppose that the two candidate echoes with the same phase reference are: as well as: In the formula, and For different candidate echo numbers.

[0051] The complex conjugate product of the two candidate echoes is: After expanding the real and imaginary parts: The remaining phase difference between two candidate echoes is determined by the following formula: In the formula, express The complex conjugate, This indicates taking the complex principal argument.

[0052] When calculating directly using I and Q data, it is written as: The remaining phase difference satisfies: The above method directly calculates the circumferential phase difference between two complex echoes. When the phase of a candidate echo is close to... The phase of another candidate echo is close to At that time, the complex conjugate product still yields a smaller actual phase difference between the two values ​​according to the principal argument, rather than obtaining a near-terminal phase difference by subtracting the ordinary values. The phase difference.

[0053] When the complex modulus is zero, the complex argument has no definite value. Candidate echoes with a modulus of zero at the same distance are not included in the comparison of the remaining phase difference with other candidate echoes. When no phase-consistent candidate echoes are formed, the candidate echoes are still processed according to the amplitude selection rules specified in step 107, and no additional amplitude weights are set.

[0054] In another specific implementation, the phase difference threshold used in step 105 is determined based on the calibration echo under the same radar operating parameters. During calibration, the position of the reflecting target is fixed, and the first step frequency, frequency step interval, number of step frequencies, sampling rate, and complex range processing method remain unchanged. The calibration echo sequentially completes physical range alignment, deterministic range-phase calculation, and range-phase compensation to obtain the remaining phase difference samples corresponding to the normal coherent echo.

[0055] Suppose that you obtain One remaining phase difference sample: Sort by numerical value from smallest to largest: In this embodiment, the 95th percentile value of the calibration sample is used as the phase difference threshold. The threshold value should not be less than... The smallest integer is denoted as The phase difference threshold is expressed as: In the formula, Indicates the phase difference threshold. This indicates the sorting number corresponding to the empirical quantile value.

[0056] As an example of threshold calculation, suppose 20 residual phase difference samples are obtained, and the sorted results are 0.028rad, 0.035rad, 0.041rad, 0.044rad, 0.050rad, 0.054rad, 0.058rad, 0.061rad, 0.065rad, 0.069rad, 0.073rad, 0.078rad, 0.082rad, 0.086rad, 0.091rad, 0.097rad, 0.103rad, 0.109rad, 0.118rad, and 0.146rad.

[0057] because Therefore The phase difference threshold is set as follows: Converted to angles: The phase difference threshold is determined by the remaining phase fluctuation range of the calibrated echo, and there is no need to set a weighting coefficient between amplitude and phase.

[0058] Let's illustrate the continuous calculation process from steps 102 to 106 using a single physical distance unit. Take the center distance of the current physical distance unit: The physical distance unit corresponds to four candidate echoes at the same distance. The distance offset, complex data before compensation, and complex data after compensation for the four candidate echoes are shown in Table 2.

[0059] Table 2. Calculation data of candidate echoes at the same distance within the same physical distance cell. The coarse range sampling position corresponding to candidate echo 1 is: The deterministic distance-phase term is: The calculation yielded: Remove complete After phase period: The phase compensation factor is: Substitute the complex values ​​of candidate echo 1 before compensation into: get: The amplitude before compensation was: The compensated amplitude is: The amplitude remained unchanged before and after phase compensation.

[0060] The range offset corresponding to candidate echo 2 is 3.1250m, and the principal phase value is: The complex data before compensation are: After complex phase compensation, we get: The complex modulus is 0.820000 and the phase is 0.350000 rad.

[0061] The range offset of candidate echo 3 is 1.5625m, and the principal phase value is: After phase compensation, the following is obtained: The corresponding amplitude is 0.790000 and the phase is 0.280000 rad.

[0062] The coarse range sampling position corresponding to candidate echo 4 is the same as the center distance of the current physical range cell, the range offset is 0, the deterministic range phase term is 0, and no change to the phase reference is required. The corresponding amplitude is 1.080000 and the phase is 1.120000 rad.

[0063] The four candidate echoes with unified phase references have a total of: A unique pair of pairs.

[0064] Substituting the compensated complex data from Table 2 into the formula for calculating the remaining phase difference, we get: The aforementioned: The following criteria are applied: The remaining phase differences between each pair of candidate echo 1, candidate echo 2, and candidate echo 3 are all less than the phase difference threshold. The remaining phase differences formed by candidate echo 4 with each of candidate echo 1, candidate echo 2, and candidate echo 3 are all greater than the phase difference threshold.

[0065] like Figure 3 As shown, the horizontal axis represents the pairwise combinations of candidate echo 1 and candidate echo 2, candidate echo 1 and candidate echo 3, candidate echo 1 and candidate echo 4, candidate echo 2 and candidate echo 3, candidate echo 2 and candidate echo 4, and candidate echo 3 and candidate echo 4, respectively. The vertical axis represents the remaining phase difference. The unit is rad; the six bars, in order of their horizontal axis values, are 0.040 rad, 0.030 rad, 0.810 rad, 0.070 rad, 0.770 rad, and 0.840 rad; the horizontal dashed line in the graph represents the phase difference threshold. The bar values ​​corresponding to candidate echo 1 and candidate echo 2, candidate echo 1 and candidate echo 3, and candidate echo 2 and candidate echo 3 are located below the phase difference threshold, while the bar values ​​corresponding to candidate echo 1 and candidate echo 4, candidate echo 2 and candidate echo 4, and candidate echo 3 and candidate echo 4 are located above the phase difference threshold.

[0066] In one screening method, the remaining phase difference between each candidate echo with a unified phase reference and other candidate echoes in the same group is determined one by one. If there is at least one remaining phase difference that is not greater than the phase difference threshold, the corresponding candidate echo is retained; if all the remaining phase differences between the candidate echoes and other candidate echoes are greater than the phase difference threshold, the corresponding candidate echo is screened out.

[0067] Based on the above conditions, candidate echo 1, candidate echo 2, and candidate echo 3 all have other candidate echoes that meet the phase difference threshold, and these three candidate echoes are determined to be phase-consistent candidate echoes. The remaining phase difference between candidate echo 4 and the other three candidate echoes all exceed 0.118 rad, therefore it does not enter the phase-consistent candidate echo range.

[0068] Another screening method is suitable for situations where there are three or more candidate echoes with unified phase references within the same physical distance cell. The number of candidate echoes that satisfy the phase difference threshold with other candidate echoes in the same group is expressed as: In the formula, when the condition is met... Set the value to 1; set the value to 0 if the condition is not met.

[0069] Based on the six remaining phase differences corresponding to Table 2, we obtain: Confirm again: This embodiment yields: when At that time, from the satisfaction Reference candidate echoes are determined from the candidate echoes. Candidate echoes 1, 2, and 3... All are 2, corresponding to amplitudes of 0.860000, 0.820000 and 0.790000 respectively. Therefore, candidate echo 1 is determined as the reference candidate echo.

[0070] Let the reference candidate echo number be The set of indexes for phase-consistent candidate echoes is represented as: In this embodiment: The remaining phase difference between candidate echo 2 and the reference candidate echo is 0.040000 rad, and the remaining phase difference between candidate echo 3 and the reference candidate echo is 0.030000 rad, both less than 0.118 rad. Therefore: The remaining phase difference between candidate echo 4 and the reference candidate echo is 0.810000 rad, which does not belong to... .

[0071] Candidate echo 1, candidate echo 2, and candidate echo 3 are entered into the amplitude comparison range in step 106. The amplitudes of phase-consistent candidate echoes are determined according to: calculate.

[0072] The candidate echo with the largest amplitude among the phase-consistent candidate echoes is denoted as... The corresponding amplitude satisfies: The amplitudes of the three phase-consistent candidate echoes are 0.860000, 0.820000, and 0.790000, respectively. Therefore: The first candidate echo becomes the extracted echo corresponding to the 1000.0000m physical distance unit.

[0073] If only the amplitude within the same physical distance unit is directly selected as the largest, the fourth candidate echo has the largest amplitude of 1.080000 among the four candidate echoes. After adopting the phase filtering method described in this embodiment, there is no phase relationship between the fourth candidate echo and the other three candidate echoes that meets the phase difference threshold, and it does not enter the amplitude comparison range of step 106. Finally, the first candidate echo is extracted. The distance phase compensation does not change the amplitude of any candidate echo. The difference in the above extraction results comes from the fact that the candidate echo discrimination criterion has been changed from a single amplitude comparison to an amplitude comparison after phase filtering.

[0074] When the same physical distance cell has only one candidate echo at the same distance, i.e. When there is no second candidate echo participating in the remaining phase difference calculation, the only candidate echo at the same distance is directly determined as the extracted echo of the corresponding physical distance unit.

[0075] when When there is only one remaining phase difference, if the remaining phase difference is not greater than the phase difference threshold, both candidate echoes are determined to be phase-consistent candidate echoes, and the echo is then extracted according to the amplitude. If the remaining phase difference is greater than the phase difference threshold, neither of the two candidate echoes can form a phase-consistent candidate echo, and after proceeding to step 107, the echo is still extracted according to the amplitude of the two candidate echoes.

[0076] When a physical distance cell has more than three candidate echoes, and the residual phase difference between all candidate echoes is greater than the phase difference threshold, each Both are 0, satisfying: At this point, the reference candidate echo is uncertain, and no phase-consistent candidate echo is formed, so proceed directly to step 107.

[0077] Step 107 uses the complex modulus of the original candidate echoes at the same distance to complete the selection. The candidate echo with the largest amplitude at the same distance is denoted as... The correspondence is as follows: The corresponding candidate echoes at the same distance are used as the extracted echoes of the current physical distance cell. If the phase relationship does not form a valid candidate range, the distance image data of the corresponding physical distance cell is still retained, and no new evaluation parameters are added.

[0078] After all physical distance cells have completed candidate echo selection, the first The extracted echo corresponding to each physical distance cell is represented as follows: One-dimensional high-resolution distance image in the 1st The amplitude at each physical distance unit is: according to: Determine the physical distance corresponding to each extracted echo, and according to Arranged from near to far This yields a one-dimensional high-resolution range image.

[0079] Range-phase compensation for candidate echoes at the same distance only changes the phase reference of the candidate echo, without changing the amplitude of the candidate echo; the remaining phase difference is used to determine the range of candidate echoes participating in amplitude comparison; when no candidate echoes with the same phase are formed, the amplitude selection branch is retained. Therefore, the formed one-dimensional high-resolution range image still maintains the original physical range grid, and the amplitude-phase screening is concentrated on the location of redundant echoes within the same physical range cell.

[0080] In one specific implementation of the system, this embodiment provides a frequency-modulated stepping radar range image extraction system based on range-coordinated redundant echo amplitude and phase filtering, used to execute the aforementioned frequency-modulated stepping radar range image extraction method based on range-coordinated redundant echo amplitude and phase filtering. The system includes a complex range data acquisition unit, a range-coordinated candidate echo group determination unit, a deterministic range phase term determination unit, a range phase compensation unit, a remaining phase difference calculation unit, a phase-consistent candidate echo determination unit, a first extracted echo determination unit, a second extracted echo determination unit, and a range image generation unit.

[0081] The complex range data acquisition unit acquires multiple sets of complex range data generated from the frequency-modulated stepping radar echoes via inverse fast Fourier transform (IFFT). Each set of complex range data corresponds to a coarse range sampling position, and the complex range data retains both the real and imaginary parts of the IFFT result. The complex range data acquisition unit associates the coarse range sampling position with the corresponding complex range data and then sends it to the candidate echo group determination unit for the same range.

[0082] The same-distance candidate echo group determination unit performs distance merging based on the physical distance corresponding to each complex distance data point. After complex distance data from different coarse distance sampling locations are mapped to a unified physical distance coordinate system, the complex data corresponding to the same physical distance unit are determined as same-distance candidate echoes. The same-distance candidate echoes corresponding to the same physical distance unit form a same-distance candidate echo group. When a physical distance unit corresponds to only one complex data point, the corresponding same-distance candidate echo group contains only one same-distance candidate echo.

[0083] The deterministic range-phase term determination unit reads the coarse range sampling position and the physical distance of the corresponding physical range unit for each candidate echo at the same range, calculates the range offset between the two, and then determines the corresponding deterministic range-phase term based on the range-phase relationship of the frequency-modulated stepping radar echo. Different candidate echoes at the same range retain their corresponding deterministic range-phase terms, enabling subsequent phase compensation to be processed separately according to the data source of each candidate echo.

[0084] The range-phase compensation unit receives a group of candidate echoes at the same range, containing multiple candidate echoes at the same range. For each candidate echo at the same range within the group, a complex phase compensation factor is formed based on the corresponding deterministic range-phase term. The candidate echo at the same range is then subjected to complex number operations with the corresponding complex phase compensation factor to obtain a candidate echo with a unified phase reference. The magnitude of the complex phase compensation factor is 1; therefore, range-phase compensation only changes the phase reference of the candidate echo, without changing the complex magnitude of the candidate echo.

[0085] The remaining phase difference calculation unit performs pairwise comparisons of phase reference candidate echoes within the same co-located candidate echo group. The relative phase between any two candidate echoes is determined using a complex conjugate product, and the absolute value of the principal argument of the complex conjugate product is taken as the remaining phase difference. Co-located candidate echo groups containing multiple candidate echoes form corresponding remaining phase difference sets, which are then sent to the phase-consistent candidate echo determination unit.

[0086] The phase-consistent candidate echo determination unit compares each remaining phase difference with a pre-determined phase difference threshold. When a candidate echo has a remaining phase difference that meets the phase difference threshold with other candidate echoes in the same group, the candidate echo with a unified phase reference is retained according to the corresponding comparison relationship, and the candidate echo that meets the screening conditions is determined as a phase-consistent candidate echo. When no candidate echo meets the phase screening conditions, the current candidate echo group at the same distance is recorded as not having formed a phase-consistent candidate echo.

[0087] The first extracted echo determination unit is used to process physical distance cells that have formed phase-consistent candidate echoes. It determines the corresponding amplitude based on the real and imaginary data of each phase-consistent candidate echo, compares the amplitudes of each phase-consistent candidate echo corresponding to the same physical distance cell, and determines the phase-consistent candidate echo with the largest amplitude as the extracted echo of the corresponding physical distance cell.

[0088] The second decimation echo determination unit is used to process physical distance units that have not formed phase-consistent candidate echoes. When a group of candidate echoes at the same distance contains only one candidate echo at the same distance, the candidate echo at the same distance is directly determined as the decimation echo. When a group of candidate echoes at the same distance contains multiple candidate echoes at the same distance but has not formed phase-consistent candidate echoes, the corresponding amplitude is determined based on the real and imaginary data of each candidate echo at the same distance within the group, and the decimation echo of the corresponding physical distance unit is determined based on the amplitude comparison result. When using the processing method described in the aforementioned specific embodiment, the candidate echo at the same distance with the largest complex modulus is determined as the decimation echo.

[0089] The range image generation unit receives the extracted echoes corresponding to each physical distance unit determined by the first and second extracted echo determination units. It establishes a distance order based on the physical distance units corresponding to the extracted echoes and arranges the extracted echoes from near to far according to their physical distance. The complex modulus of each extracted echo is used as the range image amplitude of the corresponding physical distance unit, and after arrangement, a one-dimensional high-resolution range image is obtained.

[0090] In this embodiment, the complex distance data acquisition unit and the distance-phase compensation unit complete the physical distance merging and phase reference unification of redundant echoes at the same distance. The remaining phase difference calculation unit and the phase-consistent candidate echo determination unit add phase relationship discrimination between candidate echoes at the same distance before amplitude comparison. When phase screening can form candidate results, the first extracted echo determination unit selects amplitude within the phase-consistent candidate echo range; when phase screening does not form candidate results, the second extracted echo determination unit retains the amplitude selection branch, so that the corresponding physical distance unit still has a definite extracted echo. The distance image generation unit does not change the positional relationship of each physical distance unit, but only completes the arrangement of the extracted results according to the physical distance order.

[0091] The beneficial effects of this invention are as follows: This embodiment retains the complex distance data after inverse fast Fourier transform, groups complex data from different coarse distance sampling positions that correspond to the same physical distance cell into a same-distance candidate echo group, and determines a deterministic distance phase term based on the distance offset between the coarse distance sampling position and the physical distance cell. Distance phase compensation is then applied to each candidate echo, ensuring that the candidate echoes in the same group correspond to a unified phase reference. Furthermore, the remaining phase difference between candidate echoes is calculated using the complex conjugate product, and this remaining phase difference is compared with a phase difference threshold, adding phase relationship discrimination before the traditional same-distance amplitude selection. For physical distance cells that can form phase-consistent candidate echoes, amplitudes are compared only within the phase-consistent candidate echo range, thus preventing candidate echoes with large amplitudes but significantly deviating from the phase relationship of other candidate echoes in the same group from directly entering the final extraction result. When no phase-consistent candidate echoes are formed, the amplitude selection branch based on the complex modulus is still retained, preventing the corresponding physical distance cell from lacking extracted echoes due to unmet phase screening conditions. The magnitude of the range phase compensation factor is 1, which does not change the original amplitude relationship of each candidate echo. The final one-dimensional high-resolution range image still maintains the predetermined physical range grid. Therefore, this embodiment adds a candidate discrimination criterion based on the remaining phase difference for the case where there are multiple redundant echoes in the same physical range unit without changing the original range position relationship and amplitude definition, so that the range image extraction no longer depends solely on a single amplitude comparison.

[0092] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0093] Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. Furthermore, those skilled in the art will recognize that, based on the ideas of this invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for range image extraction of frequency-modulated stepping radar based on amplitude and phase filtering of redundant echoes at the same range, characterized in that, include: Multiple sets of complex range data are obtained by inverse fast Fourier transform of the frequency modulation stepping radar echo, and each set of complex range data corresponds to a coarse range sampling position. Based on the physical distance corresponding to the complex distance data in each group, the complex data corresponding to the same physical distance unit are determined as candidate echoes of the same distance, forming a candidate echo group of the same distance; Based on the distance offset between the coarse distance sampling position corresponding to each candidate echo at the same distance and the physical distance of the corresponding physical distance unit, the deterministic distance phase term of each candidate echo at the same distance is determined; Based on the deterministic distance-phase term, distance-phase compensation is performed on the candidate echo group containing multiple candidate echoes at the same distance to obtain candidate echoes with unified phase reference. Calculate the remaining phase difference between candidate echoes of the same phase reference unified according to the same candidate echo group at the same distance; Each remaining phase difference is compared with a phase difference threshold, and the phase-consistent candidate echoes corresponding to the candidate echo group at the same distance are determined based on the comparison results. When the phase-consistent candidate echo exists, the extracted echo of the corresponding physical distance unit is determined based on the amplitude of the phase-consistent candidate echo. When no phase-consistent candidate echo is formed, the extracted echo of the corresponding physical distance unit is determined according to the amplitude of each candidate echo in the corresponding candidate echo group. The extracted echoes are arranged in the order of distance of each physical distance unit to obtain a one-dimensional high-resolution distance image.

2. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 1, is characterized in that... Multiple sets of complex range data are obtained from the frequency-modulated stepping radar echo after inverse fast Fourier transform. Each set of complex range data corresponds to a coarse range sampling position, including: The frequency-modulated step radar echoes are grouped according to the coarse range sampling positions to obtain step frequency echo sequences corresponding to each of the coarse range sampling positions. The inverse fast Fourier transform is performed on each step frequency echo sequence along the step frequency dimension, and the real and imaginary data of the inverse fast Fourier transform result are retained. The real and imaginary data corresponding to the same coarse distance sampling position are combined into a set of complex distance data.

3. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 2, is characterized in that... Based on the physical distance corresponding to each group of complex distance data, the complex data corresponding to the same physical distance unit are determined as candidate echoes of the same distance, forming a candidate echo group of the same distance, including: Based on the coarse distance sampling position corresponding to each group of complex distance data, the distance sequence number of each complex data in the group, and the distance cell interval after inverse fast Fourier transform, determine the physical distance corresponding to each complex data. The complex data corresponding to the same physical distance unit are distance aligned, and the distance-aligned complex data is determined as the candidate echo of the same distance; The candidate echoes corresponding to the same physical distance unit are combined into the candidate echo group.

4. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 3, is characterized in that... Based on the distance offset between the coarse distance sampling position corresponding to each candidate echo at the same distance and the physical distance of the corresponding physical distance unit, a deterministic range phase term for each candidate echo at the same distance is determined, including: Using the physical distance of the corresponding physical distance unit as a unified distance benchmark, the distance offset of the coarse distance sampling position corresponding to each candidate echo of the same distance relative to the unified distance benchmark is determined; Based on the range offset and the range-phase relationship of the frequency-modulated step radar echo, the deterministic range-phase term corresponding to each candidate echo at the same range is determined.

5. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 4, is characterized in that... Based on the deterministic range-phase term, range-phase compensation is performed on the candidate echo group containing multiple candidate echoes of the same range to obtain candidate echoes with a unified phase reference, including: Based on the deterministic range phase terms corresponding to each candidate echo at the same distance, complex phase compensation factors with a modulus of 1 and a phase angle that is the opposite of the corresponding deterministic range phase term are constructed respectively. Each candidate echo with the same distance is multiplied by the corresponding complex phase compensation factor so that each candidate echo in the same group with the same distance corresponds to the unified distance reference, thus obtaining candidate echoes with the unified phase reference.

6. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 5, is characterized in that... Calculating the remaining phase difference between candidate echoes with unified phase references corresponding to the same candidate echo group at the same distance includes: For any two candidate echoes with the same phase reference within the same candidate echo group, calculate the complex conjugate product of one candidate echo and the other candidate echo; The absolute value of the principal argument of the complex conjugate product is taken as the residual phase difference between the two candidate echoes of the unified phase reference, and the residual phase difference satisfies: In the formula, and They represent the first The first physical distance unit corresponding to the first The and the first Candidate echoes with unified phase reference, and For different candidate echo numbers, express The complex conjugate, Indicates taking the principal argument. This represents the remaining phase difference between two candidate echoes that are aligned with the phase reference.

7. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 6, is characterized in that... The remaining phase difference of each item is compared with a phase difference threshold, and the phase-consistent candidate echoes corresponding to the candidate echo group at the same distance are determined based on the comparison results, including: For each candidate echo with a unified phase reference, the remaining phase difference between the candidate echo with the unified phase reference and the other candidate echoes with the unified phase reference within the same group of candidate echoes at the same distance is compared with the phase difference threshold. When there is at least one remaining phase difference that is not greater than the phase difference threshold, the corresponding candidate echo of the unified phase reference is retained; when all the remaining phase differences are greater than the phase difference threshold, the corresponding candidate echo of the unified phase reference is filtered out. The candidate echoes that are consistent with the retained phase reference are determined as the phase-consistent candidate echoes.

8. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 6, is characterized in that... The remaining phase difference of each item is compared with a phase difference threshold, and the phase-consistent candidate echoes corresponding to the candidate echo group at the same distance are determined based on the comparison results, including: When the same candidate echo group at the same distance contains three or more candidate echoes with the same phase reference, count the number of remaining phase differences between each candidate echo with the same phase reference and the other candidate echoes with the same phase reference that are not greater than the phase difference threshold. Determine the maximum value among the remaining phase differences corresponding to the candidate echoes of each phase reference; When the maximum value is greater than zero, the candidate echo with the largest amplitude is selected from the candidate echoes with the same remaining phase difference as the maximum value as the reference candidate echo. The reference candidate echo and the candidate echo whose remaining phase difference with the reference candidate echo is not greater than the phase difference threshold are identified as the phase-consistent candidate echo. When the maximum value is equal to zero, it is determined that the candidate echo group with the same distance has not formed the phase-consistent candidate echo.

9. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 7, is characterized in that... When the phase-consistent candidate echo exists, the extracted echo of the corresponding physical distance unit is determined based on the amplitude of the phase-consistent candidate echo, including: The corresponding complex modulus value is determined based on the real part data and imaginary part data of each phase-consistent candidate echo, and the complex modulus value is used as the amplitude of the corresponding phase-consistent candidate echo; The amplitudes of the phase-consistent candidate echoes corresponding to the same physical distance unit are compared, and the phase-consistent candidate echo with the largest amplitude is determined as the extracted echo of the corresponding physical distance unit.

10. The method for range image extraction of frequency-modulated stepping radar based on amplitude and phase screening of redundant echoes at the same distance, as described in claim 6, is characterized in that... When no phase-consistent candidate echo is formed, the extracted echo of the corresponding physical distance unit is determined based on the amplitude of each candidate echo within the corresponding candidate echo group, including: When the same distance candidate echo group contains only one same distance candidate echo, the same distance candidate echo is determined as the extracted echo of the corresponding physical distance unit; When the same distance candidate echo group contains multiple same distance candidate echoes but does not form the phase-consistent candidate echo, the corresponding complex modulus value is determined according to the real part data and imaginary part data of each same distance candidate echo, and the same distance candidate echo with the largest complex modulus value is determined as the extracted echo of the corresponding physical distance unit.

Images