Radar signal processing device and radar signal processing program

Abstract

To provide a radar technology for extracting a target echo and removing random clutters, in which when setting a threshold for the signal strength of each cell to be processed, the target echo is reliability detected while further reducing calculation cost.SOLUTION: The present disclosure pays attention to the fact that an amplitude spectrum among the Doppler spectra of each cell to be processed is characterized in its shape in correspondence to which of a target echo and a random clutter it comes from. Therefore, an amplitude spectrum among the Doppler spectra of each cell to be processed is inverse Fourier transformed to a zero-phase signal of each cell to be processed (depending only on the shape of the amplitude spectrum). Thus, it is possible to separate between the zero-phase signal due to a random clutter and the zero-phase signal due to a target echo. Furthermore, it is possible to set the amplitude spectrum due to a random clutter to the threshold for the signal strength of each cell to be processed.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to radar techniques for extracting target echoes and removing random clutter. [Background technology]

[0002] Radar technology for extracting target echoes and removing random clutter is disclosed in Non-Patent Document 1 and the like. When the target echo to be extracted is a ship echo, an aircraft echo, a ground object echo, or the like, the random clutter to be removed is sea clutter, precipitation clutter, tree clutter, or the like. Non-Patent Document 1 particularly discloses CFAR (Constant False Alarm Rate) technology. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Kazuo Ouchi et al., "Radar Fundamentals - From Survey Radar to Synthetic Aperture Radar," Corona Publishing (2018), pp. 82-92. Summary of the Invention [Problem to be solved by the invention]

[0004] In Non-Patent Document 1, the signal strengths of cells surrounding each target cell are tallied and a threshold value for the signal strength of each target cell is set. When the signal strength of each target cell is equal to or greater than the threshold value, it is extracted as being due to a target echo. On the other hand, when the signal strength of each target cell is equal to or less than the threshold value, it is suppressed as being due to random clutter.

[0005] The simplest threshold setting method is to use the average signal strength of neighboring cells (CA (Cell Averaging)-CFAR). However, when a target exists in a neighboring cell, the threshold is set high, making it difficult to detect the target echo.

[0006] To avoid this, one method for setting a threshold value is to use the median value of the signal strengths of neighboring cells (OS (Order Statistic)-CFAR). However, this increases the computational cost because the signal strengths of neighboring cells must be sorted in ascending or descending order.

[0007] Therefore, in order to solve the above-mentioned problems, an object of the present disclosure is to reliably detect target echoes while further reducing the calculation cost when setting a threshold value for the signal strength of each processing target cell in a radar technology that extracts target echoes and removes random clutter. [Means for solving the problem]

[0008] To solve the above problem, we focus on the fact that the amplitude spectrum of the Doppler spectrum of each processing target cell has a characteristic shape depending on whether it is due to a target echo or random clutter. Here, the amplitude spectrum due to a target echo has a shape that spreads smoothly around a certain Doppler velocity, while the amplitude spectrum due to random clutter has a shape that spreads in a disordered manner around a certain Doppler velocity.

[0009] Therefore, the amplitude spectrum of the Doppler spectrum of each processing cell is inverse Fourier transformed to the zero-phase signal of each processing cell (which depends only on the shape of the amplitude spectrum). Here, the zero-phase signal due to the target echo is concentrated only in the time domain near the origin. On the other hand, the zero-phase signal due to random clutter remains in the time domain other than the vicinity of the origin.

[0010] Therefore, the zero-phase signal due to random clutter can be separated from the zero-phase signal due to the target echo.The zero-phase signal due to random clutter is then Fourier transformed into an amplitude spectrum due to random clutter.The amplitude spectrum due to random clutter can then be set as the signal strength threshold for each processing cell.

[0011] Specifically, the present disclosure provides a radar signal processing device that extracts target echoes and removes random clutter, the device including a Doppler spectrum calculation unit that performs a Fourier transform on radar observation signals for a plurality of sweeps of each processing target cell and calculates a Doppler spectrum of the processing target cell, a zero-phase signal calculation unit that performs an inverse Fourier transform on an amplitude spectrum of the Doppler spectrum of the processing target cell and calculates a zero-phase signal of the processing target cell, a zero-phase signal extraction unit that applies a window function to the zero-phase signal of each processing target cell, suppresses a time region near an origin where the signal strength of the target echo is concentrated, and extracts a time region other than the origin where the signal strength of the random clutter remains, and a zero-phase signal extraction unit that applies a window function to the zero-phase signal of each processing target cell, suppresses a time region near an origin where the signal strength of the random clutter remains, and extracts a time region other than the time region near the origin where the signal strength of the random clutter remains. a Doppler spectrum restoration unit that performs a Fourier transform on the zero-phase signal of the signal obtained by the Doppler spectrum calculation unit to suppress the signal intensity of the target echo and extract the signal intensity of the random clutter, and then restores an amplitude spectrum from the Doppler spectrum of each of the processing target cells; and a threshold processing unit that sets the amplitude spectrum from the Doppler spectrum of each of the processing target cells restored by the Doppler spectrum restoration unit as a threshold, and extracts the target echo whose signal intensity is greater than the threshold, and removes the random clutter whose signal intensity is smaller than the threshold.

[0012] According to this configuration, the signal strength threshold of each processing target cell is set by only performing a Fourier transform within each processing target cell, without aggregating the signal strengths of the cells surrounding each processing target cell, so that the target echo can be reliably detected while further reducing the calculation cost.

[0013] The present disclosure also provides a radar signal processing device, wherein the Doppler spectrum calculation unit calculates an amplitude spectrum of the Doppler spectrum of each processing target cell as a first-power amplitude spectrum.

[0014] With this configuration, the zero-phase signals of the amplitude spectrum of the target echo, which is the first power, are concentrated only in the time domain near the origin, compared with the zero-phase signals of the amplitude spectrum of the target echo, which is the squared power. Therefore, by expanding the time width of the window function that suppresses the zero-phase signals of the target echo and extracts the zero-phase signals of the random clutter, it is possible to extract almost all of the zero-phase signals of the random clutter, restore the amplitude spectrum of the random clutter with high accuracy, and set the signal strength threshold of each processing cell with high accuracy.

[0015] The present disclosure also provides a radar signal processing device, characterized in that the zero-phase signal extraction unit sets a time width of the window function that takes into account a velocity width of the target echo that is larger than the velocity resolution of the radar signal processing device.

[0016] With this configuration, zero-phase signals in the amplitude spectrum due to target echoes wider than the velocity resolution are concentrated only in the time domain near the origin, compared to zero-phase signals in the amplitude spectrum due to target echoes close to the velocity resolution. Therefore, by expanding the time width of the window function that suppresses zero-phase signals due to target echoes and extracts zero-phase signals due to random clutter, it is possible to extract almost all zero-phase signals due to random clutter, accurately restore the amplitude spectrum due to random clutter, and accurately set the signal strength threshold for each processing cell.

[0017] The present disclosure also provides a radar signal processing device, characterized in that the zero-phase signal extraction unit corrects power loss after application of the window function so that the Doppler amplitude spectrum of each processing target cell restored by the Doppler spectrum restoration unit is equivalent in signal strength of the random clutter to the Doppler amplitude spectrum of each processing target cell calculated by the Doppler spectrum calculation unit.

[0018] According to this configuration, by correcting the power loss in advance at the stage of applying the window function, it is possible to reliably extract the target echo and set the signal strength threshold of each processing target cell with high accuracy so as to reliably remove random clutter.

[0019] The present disclosure also provides a radar signal processing device, characterized in that the threshold processing unit employs a Doppler amplitude spectrum of each processing target cell restored by the Doppler spectrum restoration unit to which a predetermined dB has been added, so that the Doppler amplitude spectrum of each processing target cell restored by the Doppler spectrum restoration unit is larger in terms of signal strength of the random clutter than the Doppler amplitude spectrum of each processing target cell calculated by the Doppler spectrum calculation unit.

[0020] According to this configuration, by further adding a predetermined dB at the stage of setting the threshold, it is possible to set the threshold of the signal strength of each processing target cell with high accuracy so as to reliably extract target echoes and reliably remove random clutter.

[0021] The present disclosure also provides a radar signal processing program for causing a computer to sequentially execute the processing steps performed by the processing units included in the above-described radar signal processing device.

[0022] According to this configuration, it is possible to provide a program having the above-described effects. [Effects of the Invention]

[0023] In this way, the present disclosure enables reliable detection of target echoes while further reducing the computational cost when setting a signal strength threshold for each processing target cell in a radar technology that extracts target echoes and removes random clutter. [Brief explanation of the drawings]

[0024] [Figure 1]FIG. 1 is a diagram illustrating a configuration of a radar signal processing device according to the present disclosure. [Figure 2] FIG. 2 is a diagram showing a procedure of a radar signal processing program according to the present disclosure. [Figure 3] FIG. 10 is a diagram showing the results of actual measurement of the Doppler spectrum of the present disclosure. [Figure 4] FIG. 2 is a diagram illustrating an amplitude spectrum of a radar observation signal according to the present disclosure. [Figure 5] FIG. 10 illustrates an amplitude spectrum zero phase signal of the present disclosure. [Figure 6] FIG. 10 is a diagram illustrating a window function of the zero phase signal extraction unit of the present disclosure. [Figure 7] FIG. 10 illustrates a zero-phase signal after application of a window function according to the present disclosure. [Figure 8] FIG. 10 illustrates the amplitude spectrum of a zero phase signal of the present disclosure. [Figure 9] FIG. 10 is a diagram illustrating a comparison result between amplitude spectra according to the present disclosure. [Figure 10] FIG. 10 is a diagram illustrating a calculation process of an amplitude spectrum according to the present disclosure. [Figure 11] FIG. 10 is a diagram illustrating a process for setting a target echo velocity width according to the present disclosure. [Figure 12] FIG. 10 is a diagram illustrating a window function power loss correction process according to the present disclosure. [Figure 13] FIG. 10 is a diagram illustrating the addition process of amplitude spectra according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0025]

[0023] The following embodiments of the present disclosure will be described with reference to the accompanying drawings. The embodiments described below are examples of implementation of the present disclosure, and the present disclosure is not limited to the following embodiments.

[0026] (Overview of radar signal processing device of the present disclosure) The configuration of the radar signal processing device of the present disclosure is shown in Fig. 1. The procedure of the radar signal processing program of the present disclosure is shown in Fig. 2. The radar signal processing device R is composed of a Doppler spectrum calculation unit 1, a zero-phase signal calculation unit 2, a zero-phase signal extraction unit 3, a Doppler spectrum restoration unit 4, a threshold processing unit 5, and a maximum value output unit 6. The radar signal processing device R can be realized by installing the radar signal processing program shown in Fig. 2 on a computer.

[0027] The results of actual measurements of the Doppler spectrum of this disclosure are shown in Figure 3. When the target echo to be extracted is a ship echo, an aircraft echo, a ground object echo, or the like, the random clutter to be removed is sea clutter, precipitation clutter, tree clutter, or the like, respectively.

[0028] The amplitude spectrum of the Doppler spectrum of each processing cell has a characteristic shape depending on whether it is due to a target echo or random clutter. Here, the amplitude spectrum due to a target echo has a shape that spreads smoothly around a certain Doppler velocity depending on the almost unique velocity of the target and the velocity resolution of the radar. On the other hand, the amplitude spectrum due to random clutter has a shape that spreads disorderly around a certain Doppler velocity depending on the almost uniform velocity of the clutter as a whole and the random velocity of the clutter individually.

[0029] Therefore, the amplitude spectrum of the Doppler spectrum of each processing cell is inverse Fourier transformed to the zero-phase signal of each processing cell (which depends only on the shape of the amplitude spectrum). Here, the zero-phase signal due to the target echo is concentrated only in the time domain near the origin. On the other hand, the zero-phase signal due to random clutter remains in the time domain other than the vicinity of the origin.

[0030] Therefore, the zero-phase signal due to random clutter can be separated from the zero-phase signal due to the target echo.The zero-phase signal due to random clutter is then Fourier transformed into an amplitude spectrum due to random clutter.The amplitude spectrum due to random clutter can then be set as the signal strength threshold for each processing cell.

[0031] (Specific example of radar signal processing procedure of the present disclosure) The amplitude spectrum of the radar observation signal of the present disclosure is shown in Fig. 4. The Doppler spectrum calculation unit 1 performs a Fourier transform on the radar observation signal x(t) for multiple sweeps of each processing target cell, and calculates the Doppler spectrum X(f) (amplitude spectrum |X(f)| 2 and the phase spectrum ∠X(f) are calculated (step S1).

[0032] In the left column of Fig. 4, the amplitude spectrum |X(f)| due to the target echo is 2 has a shape that spreads smoothly around a certain Doppler velocity. In the middle column of Figure 4, the amplitude spectrum |X(f)| due to random clutter 2 has a shape that spreads around a certain Doppler velocity. In the right column of Fig. 4, the amplitude spectrum |X(f)| due to the target echo 2 and the amplitude spectrum due to random clutter |X(f)| 2 and are superimposed.

[0033] The zero-phase signal of the amplitude spectrum of the present disclosure is shown in Fig. 5. The zero-phase signal calculation unit 2 calculates the amplitude spectrum |X(f)| of the Doppler spectrum X(f) of each processing target cell. 2 is subjected to an inverse Fourier transform to calculate the zero phase signal x0(t) of each processing target cell (step S2).

[0034] In the left column of Figure 5, the zero-phase signal x0(t) due to the target echo is concentrated only in the time domain near the origin t = 0 (including the turning point t = 64). In the middle column of Figure 5, the zero-phase signal x0(t) due to random clutter remains in time domains other than the vicinity of the origin t = 0 (including the turning point t = 64). In the right column of Figure 5, the zero-phase signal x0(t) due to the target echo and the zero-phase signal x0(t) due to random clutter are superimposed.

[0035] Here, the amplitude spectrum of the target echo |X(f)| 2 has a smoothly spreading shape (corresponding to the DC-near component in the time domain), and as a result of its inverse Fourier transform, the zero-phase signal x0(t) due to the target echo is concentrated only in the time domain near the origin t = 0 (corresponding to the DC-near component in the frequency domain). On the other hand, the amplitude spectrum |X(f)| due to random clutter 2 has a disordered and spreading shape (corresponding to high-frequency components in the time domain), and as a result of its inverse Fourier transform, the zero-phase signal x0(t) due to random clutter remains in the time domain outside the vicinity of the origin t=0 (corresponding to high-frequency components in the frequency domain).

[0036] Or, the amplitude spectrum of the target echo |X(f)| 2 is expressed by a single Gaussian function (the peak width of a Gaussian function is wide), and as a result of its inverse Fourier transform, the zero-phase signal x0(t) due to the target echo is concentrated only in the time domain near the origin t = 0 (corresponding to the DC-near component in the frequency domain). On the other hand, the amplitude spectrum |X(f)| due to random clutter 2 is expressed as a composition of multiple Gaussian functions (the peak widths of Gaussian functions vary), and as a result of the inverse Fourier transform, the zero-phase signal x0(t) due to random clutter remains in the time domain outside the vicinity of the origin t=0 (corresponding to high-frequency components in the frequency domain).

[0037] The window function of the zero-phase signal extraction unit of the present disclosure is shown in Figure 6. The zero-phase signal after application of the window function of the present disclosure is shown in Figure 7. The zero-phase signal extraction unit 3 applies the window function w(t) to the zero-phase signal x0(t) of each processing target cell, suppresses the time region near the origin t = 0 (including the turning point t = 64) where the signal strength of the target echo is concentrated, extracts the time region other than the vicinity of the origin t = 0 (including the turning point t = 64) where the signal strength of random clutter remains, and extracts the zero-phase signal x0'(t) of each processing target cell (step S3).

[0038] In FIG. 6, the window function w(t) is, for example, a Hanning window function, which suppresses the time region around the origin t=0 (including the turn-around point t=64) where the signal strength of the target echo is concentrated in the zero-phase signal x0(t) of each processing cell, and extracts the time region other than the region around the origin t=0 (including the turn-around point t=64) where the signal strength of random clutter remains.

[0039] In the left column of FIG. 7, the zero-phase signal x0'(t) due to the target echo is suppressed in the time domain near the origin t=0 (including the turning point t=64) where the signal strength of the target echo is concentrated. In the middle column of FIG. 7, the zero-phase signal x0'(t) due to random clutter is extracted in the time domain other than the vicinity of the origin t=0 (including the turning point t=64) where the signal strength of the random clutter remains. In the right column of FIG. 7, the zero-phase signal x0'(t) due to the target echo and the zero-phase signal x0'(t) due to random clutter are superimposed. Here, in the middle and right columns of FIG. 7, the window function power loss correction process shown in the bottom column of FIG. 12 is executed.

[0040] The amplitude spectrum of the zero-phase signal of the present disclosure is shown in Fig. 8. The Doppler spectrum restoration unit 4 performs a Fourier transform on the zero-phase signal x0'(t) of each processing target cell extracted by the zero-phase signal extraction unit 3, suppressing the signal strength of the target echo and extracting the signal strength of random clutter, and then converting the amplitude spectrum |X0(f)| of the Doppler spectrum X0(f) of each processing target cell into 2 is restored (step S4).

[0041] In the left column of Fig. 8, the amplitude spectrum |X0(f)| due to the target echo is 2 In the middle column of Fig. 8, the amplitude spectrum |X0(f)| due to random clutter is suppressed around a certain Doppler velocity. 2 The signal strength of random clutter remains around a certain Doppler velocity. In the right column of Fig. 8, the amplitude spectrum |X0(f)| due to the target echo is 2 and the amplitude spectrum due to random clutter |X0(f)| 2 and are superimposed.

[0042] The comparison result between amplitude spectra according to the present disclosure is shown in Fig. 9. The threshold processing unit 5 calculates the amplitude spectrum |X(f)| of the Doppler spectrum X(f) of each processing target cell calculated by the Doppler spectrum calculation unit 1. 2 In the Doppler spectrum X0(f) of each processing target cell restored by the Doppler spectrum restoration unit 4, the amplitude spectrum |X0(f)| 2 is set as a threshold value, target echoes whose signal strength is greater than the threshold value are extracted, and random clutter whose signal strength is smaller than the threshold value is removed (step S5).

[0043] The left column of Fig. 9 shows the amplitude spectrum |X(f)| due to the target echo and random clutter shown in the right column of Fig. 4. 2 The middle column of Fig. 9 shows the amplitude spectrum |X0(f)| due to random clutter shown in the right column of Fig. 8. 2 In the right column of FIG. 9, the threshold processing unit 5 calculates the amplitude spectrum |X0(f)| due to random clutter. 2 is used as the threshold value, and the amplitude spectrum of the target echo |X t (f)| 2 Then, the maximum value output unit 6 extracts the amplitude spectrum |X t (f)| 2 From the maximum value of the amplitude spectrum |X m (f)| 2(Step S6) Here, in the right column of Fig. 9, the addition process of the amplitude spectrum shown in the bottom column of Fig. 13 is executed.

[0044] In this way, the signal strength threshold of each processing target cell is set by only performing a Fourier transform within each processing target cell, without aggregating the signal strengths of the cells surrounding each processing target cell, so that target echoes can be reliably detected while further reducing the calculation cost.

[0045] (Improvements to the radar signal processing procedure of the present disclosure) The amplitude spectrum calculation process of the present disclosure is shown in Fig. 10. The Doppler spectrum calculation unit 1 calculates the amplitude spectrum |X(f)| in the form of a first power as the amplitude spectrum of the Doppler spectrum X(f) of each processing target cell (step S1).

[0046] In the upper panel of Fig. 10, the squared amplitude spectrum |X(f)| due to the target echo 2 has a smaller velocity width than the squared amplitude spectrum |X(f)| due to the target echo. Therefore, the squared amplitude spectrum |X(f)| due to the target echo 2 The zero-phase signal x0(t) of the amplitude spectrum |X(f)| due to the target echo remains in the time domain other than the vicinity of the origin, compared to the zero-phase signal x0(t) of the amplitude spectrum |X(f)| due to the target echo, which remains in the first power form. Furthermore, the time width of the window function w(t) cannot be expanded, and some of the zero-phase signal x0(t) due to random clutter cannot be extracted, and the amplitude spectrum |X0(f)| due to random clutter 2 Therefore, it may not be possible to accurately restore the signal strength threshold of each processing target cell.

[0047] In the bottom panel of Fig. 10, the squared amplitude spectrum |X(f)| of the target echo is compared with the squared amplitude spectrum |X(f)| of the target echo. 2 Therefore, the zero-phase signal x0(t) of the squared amplitude spectrum |X(f)| due to the target echo is larger than the zero-phase signal x0(t) of the squared amplitude spectrum |X(f)| due to the target echo. 2Compared to the zero-phase signal x0(t) of the random clutter signal, the signal intensity of the random clutter signal x0(t) is concentrated only in the time domain near the origin. If the signal-to-noise ratio is high enough, the time width of the window function w(t) can be expanded, which makes it possible to extract almost all of the zero-phase signal x0(t) due to random clutter, and to accurately restore the amplitude spectrum |X0(f)| due to random clutter, allowing the signal intensity threshold of each processing cell to be set with high accuracy.

[0048] The process of setting the target echo velocity width according to the present disclosure is shown in Fig. 11. The zero-phase signal extraction unit 3 sets the time width of the window function w(t) in consideration of the velocity width of the target echo, which is larger than the velocity resolution of the radar signal processing device R (step S3).

[0049] In the upper panel of Fig. 11, the amplitude spectrum |X(f)| due to the target echo close to the velocity resolution 2 is the amplitude spectrum |X(f)| due to the target echo wider than the velocity resolution. 2 Therefore, the amplitude spectrum |X(f)| due to the target echo, which is close to the velocity resolution, is 2 The zero-phase signal x0(t) is the amplitude spectrum |X(f)| due to the target echo, which is wider than the velocity resolution. 2 Compared with the zero-phase signal x0(t) of the random clutter, it remains in the time domain other than the vicinity of the origin. Furthermore, the time width of the window function w(t) cannot be expanded, and the zero-phase signal x0(t) due to the random clutter cannot be partially extracted, and the amplitude spectrum |X0(f)| due to the random clutter 2 cannot be restored with high accuracy, and the signal strength threshold of each processing target cell cannot be set with high accuracy.

[0050] In the lower panel of Fig. 11, the amplitude spectrum |X(f)| due to the target echo wider than the velocity resolution 2 is the amplitude spectrum |X(f)| due to the target echo close to the velocity resolution. 2 Therefore, the amplitude spectrum |X(f)| due to the target echo, which is wider than the velocity resolution, is 2 The zero-phase signal x0(t) is the amplitude spectrum |X(f)| due to the target echo close to the velocity resolution. 2Compared with the zero-phase signal x0(t) in (1), it is concentrated only in the time domain near the origin. In addition, the time width of the window function w(t) can be expanded, making it possible to extract almost all of the zero-phase signal x0(t) due to random clutter, restoring the amplitude spectrum |X0(f)| due to random clutter with high accuracy, and setting the signal strength threshold for each processing cell with high accuracy.

[0051] The window function power loss correction process of the present disclosure is shown in Fig. 12. The Doppler amplitude spectrum |X0(f)| of each processing target cell restored by the Doppler spectrum restoration unit 4 is 2 is the Doppler amplitude spectrum |X(f)| of each processing target cell calculated by the Doppler spectrum calculation unit 1. 2 The zero-phase signal extractor 3 corrects the power loss after application of the window function w(t) so that the signal strength of the random clutter is equivalent to that of the signal obtained by the window function w(t) (step S3).

[0052] In the upper part of Fig. 12, the power loss is not corrected in advance at the stage of applying the window function w(t). Therefore, the Doppler amplitude spectrum |X0(f)| 2 is the Doppler amplitude spectrum |X(f)| 2 Therefore, it is not possible to set the threshold value of the signal strength of each processing target cell with high precision so as to reliably extract the target echo and reliably remove the random clutter.

[0053] In the lower part of Fig. 12, the power loss is corrected in advance at the stage of applying the window function w(t). Therefore, the Doppler amplitude spectrum |X0(f)| 2 is the Doppler amplitude spectrum |X(f)| 212 and 13, the signal strength of random clutter is equivalent. Therefore, the signal strength threshold of each processing target cell can be set with high accuracy so that the target echo is reliably extracted and the random clutter is reliably removed. The correction coefficient for the power loss after application of the window function w(t) can be set based on empirical rules, etc. Furthermore, the signal strength threshold of each processing target cell can be set with high accuracy by combining or independently using the window function power loss correction process shown in FIG. 12 and the amplitude spectrum addition process shown in FIG. 13.

[0054] The addition process of the amplitude spectrum according to the present disclosure is shown in Fig. 13. The Doppler amplitude spectrum |X0(f)| of each processing target cell restored by the Doppler spectrum restoration unit 4 is 2 is the Doppler amplitude spectrum |X(f)| of each processing target cell calculated by the Doppler spectrum calculation unit 1. 2 The threshold processing unit 5 calculates the Doppler amplitude spectrum |X0(f)| of each processing target cell restored by the Doppler spectrum restoration unit 4 so that the signal strength of random clutter is larger than that of the Doppler spectrum restoration unit 4. 2 A predetermined dB is further added to the result (step S5).

[0055] In the upper part of Fig. 13, a predetermined dB is not added at the stage of setting the threshold. Therefore, the Doppler amplitude spectrum |X0(f)| 2 is the Doppler amplitude spectrum |X(f)| 2 The signal strength of random clutter is equivalent to that of the target echo |X t (f)| 2 , |X m (f)| 2 However, it is not possible to set the signal strength threshold of each processing target cell with high precision so as to reliably extract the target cell and reliably remove random clutter.

[0056] In the lower part of Fig. 13, a predetermined dB is further added at the stage of setting the threshold. Therefore, the Doppler amplitude spectrum |X0(f)| 2 is the Doppler amplitude spectrum |X(f)| 2Compared to the signal strength of random clutter, it is large. And the target echo |X t (f)| 2 , |X m (f)| 2 The signal strength threshold of each processing target cell can be set with high accuracy so that random clutter can be reliably removed while reliably extracting the signal strength of the target cell. The predetermined dB addition amount when setting the threshold may be set based on empirical rules, etc. Furthermore, the signal strength threshold of each processing target cell can be set with high accuracy by combining or independently using the amplitude spectrum addition process shown in FIG. 13 and the window function power loss correction process shown in FIG. 12. [Industrial Applicability]

[0057] The radar signal processing device and radar signal processing program of the present disclosure can extract target echoes such as ship echoes, aircraft echoes, and ground object echoes after removing random clutter such as sea clutter, rain clutter, and tree clutter. [Explanation of symbols]

[0058] R: Radar signal processing device 1: Doppler spectrum calculation section 2: Zero phase signal calculation section 3: Zero phase signal extraction section 4: Doppler spectrum restoration section 5: Threshold processing section 6: Maximum value output section

Claims

1. A radar signal processing device that extracts target echoes and removes random clutter, a Doppler spectrum calculation unit that performs a Fourier transform on radar observation signals for a plurality of sweeps of each processing target cell and calculates a Doppler spectrum of the processing target cell; a zero-phase signal calculation unit that performs an inverse Fourier transform on an amplitude spectrum of the Doppler spectrum of each of the processing target cells and calculates a zero-phase signal of each of the processing target cells; a zero-phase signal extracting unit that applies a window function to the zero-phase signal of each processing target cell, suppresses a time region near an origin where the signal strength of the target echo is concentrated, and extracts a time region other than the time region near the origin where the signal strength of the random clutter remains; a Doppler spectrum restoration unit that performs a Fourier transform on the zero phase signal of each processing target cell extracted by the zero phase signal extraction unit, suppresses the signal intensity of the target echo, extracts the signal intensity of the random clutter, and then restores an amplitude spectrum of the Doppler spectrum of each processing target cell; a threshold processing unit that, in an amplitude spectrum of the Doppler spectrum of each processing target cell calculated by the Doppler spectrum calculation unit and restored by the Doppler spectrum restoration unit, sets the amplitude spectrum of the Doppler spectrum of each processing target cell calculated by the Doppler spectrum calculation unit as a threshold, extracts the target echo whose signal intensity is greater than the threshold, and removes the random clutter whose signal intensity is smaller than the threshold; A radar signal processing device comprising:

2. The Doppler spectrum calculation unit calculates an amplitude spectrum of the Doppler spectrum of each processing target cell as a first-power amplitude spectrum.

2. The radar signal processing device according to claim 1, wherein:

3. The zero-phase signal extraction unit sets a time width of the window function taking into consideration a velocity width of the target echo that is larger than the velocity resolution of the radar signal processing device.

3. The radar signal processing device according to claim 1, wherein:

4. The zero-phase signal extraction unit corrects power loss after application of the window function so that the Doppler amplitude spectrum of each processing target cell restored by the Doppler spectrum restoration unit is equivalent to the Doppler amplitude spectrum of each processing target cell calculated by the Doppler spectrum calculation unit in terms of signal strength of the random clutter.

4. The radar signal processing device according to claim 1, wherein the first and second signals are input to the radar.

5. The threshold processing unit further adds a predetermined dB to the Doppler amplitude spectrum of each processing target cell restored by the Doppler spectrum restoration unit so that the Doppler amplitude spectrum of each processing target cell restored by the Doppler spectrum restoration unit is larger in terms of the signal strength of the random clutter than the Doppler amplitude spectrum of each processing target cell calculated by the Doppler spectrum calculation unit.

5. The radar signal processing device according to claim 1, wherein the first and second signals are input to the radar.

6. 6. A radar signal processing program for causing a computer to sequentially execute each processing step performed by each processing unit included in the radar signal processing device according to claim 1.

Images