Signal processing device, signal processing method, and signal processing system

Abstract

A signal processing device (3) is configured to comprise: an approximate curve calculation unit (11) that calculates an approximate curve of an actual slant range change, which is a curve indicating the actual distance change with the passage of time between an observation object and a platform (1) equipped with a radar (2) that radiates a radio wave toward the observation object and then receives the radio wave reflected by the observation object; and a division unit (12) that divides each of the actual slant range change and a reception signal of the radio wave by the radar (2) into a plurality of mutually different time zones, and calculates each approximate slant range change by approximating each actual slant range change after division. The signal processing device (3) is also provided with: a compensation amount calculation unit (13) that calculates, on the basis of the difference between each actual slant range change after division by the division unit (12) and each approximate slant range change calculated by the division unit (12), a compensation amount for compensating for the phase of each reception signal after division by the division unit (12); a compensation unit (14) that uses each compensation amount calculated by the compensation amount calculation unit (13) to compensate for the phase of each reception signal after division by the division unit (12); and a divided image synthesis unit (19) that synthesizes a plurality of images reproduced on the basis of each reception signal after phase compensation by the compensation unit (14).

Description

Signal processing device, signal processing method, and signal processing system

[0001] This disclosure relates to a signal processing device, a signal processing method, and a signal processing system.

[0002] There is a signal processing device that is mounted on a moving object, emits radio waves toward an object to be observed, and then receives the radio waves after they have been reflected by the object to reconstruct an image based on the received signal of the radar. As such a signal processing device, for example, Patent Document 1 discloses a signal processing device that uses an approximation curve of the actual slant range change, which is a curve that shows the actual change in distance over time between the platform on which the radar is mounted and the object to be observed. The signal processing device disclosed in Patent Document 1 compensates for the phase of the radar's received signal over the entire observation area based on the difference between the actual slant range change at a certain reference point within the observation area and the calculated approximation curve, and then reconstructs an image based on the received signal after phase compensation.

[0003] Japanese Patent Publication No. 2012-127751

[0004] The visibility of the observed object can be improved, for example, by performing multi-look processing. Multi-look processing is a process that divides the composite aperture into several independent sub-apertures, reconstructs an image for each sub-aperture, and then combines the multiple reconstructed images. The radar observation time in multi-look processing is the composite aperture time, which includes the times of multiple sub-apertures. To achieve multi-look processing while maintaining resolution, the radar observation time needs to be several times longer than the observation time without multi-look processing. The signal processing device disclosed in Patent Document 1 calculates a single approximation curve based on the actual slant range change over the entire radar observation time. The difference between the approximation curve and the actual slant range change is compensated based on a certain point within the observation area, but since the amount of compensation is uniform within the observation area, the compensation error increases the further away the area is from the reference point, and the image reconstruction accuracy deteriorates. When this signal processing device is applied to long-term observation data that assumes multi-look processing, the discrepancy between the actual slant range change and the approximation curve becomes very large. Therefore, even when compensating for the difference between the approximation curve and the actual slant range change using a point within the observation area as a reference, there was a problem in that the compensation error would become large in areas far from the reference point, degrading the accuracy of image reproduction.

[0005] This disclosure was made to solve the above-mentioned problems, and aims to provide a signal processing device that can suppress the deterioration of image reproduction accuracy even in situations where the approximation error becomes large when approximating the actual slant range change, such as long-term observation data, using a single curve.

[0006] The signal processing device according to this disclosure includes: an approximation curve calculation unit that calculates an approximation curve of the actual slant range change, which is a curve showing the actual distance change over time between a platform equipped with a radar that emits radio waves toward an observation target and receives radio waves after they have been reflected by the observation target, and the observation target; a division unit that divides the actual slant range change and the received signal of the radio waves by the radar into multiple different time periods and calculates each approximate slant range change by approximating the actual slant range change after division. The signal processing device also includes: a compensation amount calculation unit that calculates a compensation amount to compensate for the phase of each received signal after division by the division unit based on the difference between each actual slant range change after division by the division unit and each approximate slant range change calculated by the division unit; a compensation unit that compensates for the phase of each received signal after division by the division unit using each compensation amount calculated by the compensation amount calculation unit; and a division image synthesis unit that synthesizes multiple images reconstructed based on each received signal after phase compensation by the compensation unit.

[0007] According to this disclosure, even in situations where approximating the actual slant range change using a single curve, such as in long-term observation data, would result in a large approximation error, the degradation of image reproduction accuracy can be suppressed.

[0008] This is a diagram showing a signal processing system including a signal processing device 3 according to Embodiment 1. This is a hardware diagram showing the hardware of the signal processing device 3 according to Embodiment 1. This is a hardware diagram of a computer when the signal processing device 3 is implemented by software or firmware, etc. This is a flowchart showing the signal processing method, which is the processing procedure of the signal processing device 3. Figure 5A is an explanatory diagram showing the observation geometry of the region including the observed target Tgt when the platform 1 is moving on a linear trajectory, and Figure 5B is an explanatory diagram showing the observation geometry of the region including the observed target Tgt when the platform 1 is moving on a non-linear trajectory. This is an explanatory diagram showing an example of each in the approximate curve AC and the actual slant range change RC. With M=2, the actual slant range change dRC after division ｍand the change in approximate slant range after division dAC ｍ This is an explanatory diagram showing an example of each of the above. This is an explanatory diagram showing multi-look processing.

[0009] To provide a more detailed explanation of this disclosure, the forms for implementing this disclosure will be described below with reference to the attached drawings.

[0010] Embodiment 1. Figure 1 is a configuration diagram showing a signal processing system including a signal processing device 3 according to Embodiment 1. Figure 2 is a hardware configuration diagram showing the hardware of the signal processing device 3 according to Embodiment 1. In Figure 1, platform 1 is a mobile object such as a satellite, airplane, helicopter, flying car, or vehicle. Platform 1 is equipped with radar 2. Radar 2 is a synthetic aperture radar, which emits radio waves toward a region including the observation target Tgt, and then receives the radio waves after reflection by the region including the observation target Tgt. Radar 2 outputs a received radio wave signal S(η) to the signal processing device 3. η is the azimuth time.

[0011] The signal processing device 3 includes an approximation curve calculation unit 11, a division unit 12, a compensation amount calculation unit 13, a compensation unit 14, a range compression unit 15, a time distance calculation unit 16, an effective velocity calculation unit 17, an azimuth compression unit 18, and a division image synthesis unit 19. The approximation curve calculation unit 11 is implemented, for example, by the approximation curve calculation circuit 31 shown in Figure 2. The approximation curve calculation unit 11 calculates an approximation curve AC of the actual slant range change RC, which is a curve showing the actual distance change between the platform 1 and the observed object Tgt over time. Specifically, the approximation curve calculation unit 11 calculates the slant range, which is the distance between the platform 1 and the observed object Tgt, and calculates the approximation curve AC by approximating the slant range at multiple times with a hyperbola. The approximation curve calculation unit 11 outputs to each of the time distance calculation units 16.

[0012] The splitting unit 12 is implemented, for example, by the splitting circuit 32 shown in Figure 2. The splitting unit 12 acquires the radio wave reception signal S(η) from the radar 2. The splitting unit 12 calculates the closest approach time η from the time distance calculation unit 16. ０The information indicating and the closest distance R ０ Obtain the information indicating and. The dividing unit 12 divides each of the actual slant range change RC and the reception signal S(η) of the radio wave by the radar 2 into M different time zones TZ １ ~TZ Ｍ M is an integer of 2 or more. The time lengths of the M time zones TZ １ ~TZ Ｍ may all be the same time length, or may be different time lengths from each other. When the time lengths of the M time zones TZ １ ~TZ Ｍ are different from each other, for example, among the observation times of the reception signal S(η), the time lengths of the time zones at the observation start time and the observation end time may be divided so as to be shorter than the time length of the time zone at the observation middle time. The dividing unit 12 uses the closest time η ０ and the closest distance R ０ to approximate the actual slant range change dRC ｍ after division, thereby calculating the approximate slant range change dAC ｍ after division. The dividing unit 12 outputs the information indicating the actual slant range change dRC ｍ after division, the information indicating the approximate slant range change dAC ｍ after division, and the reception signal dS ｍ (η) to the compensation amount calculation unit 13. m = 1,..., M. Also, the dividing unit 12 outputs the information indicating the actual slant range change dRC ｍ after division and the information indicating the approximate slant range change dAC ｍ after division to the effective speed calculation unit 17. Further, the dividing unit 12 outputs the reception signal dS ｍ (η) after division to the compensation unit 14.

[0013] The compensation amount calculation unit 13 is realized by, for example, the compensation amount calculation circuit 33 shown in FIG. 2. The compensation amount calculation unit 13 receives from the dividing unit 12 the information indicating the actual slant range change dRC ｍ (m = 1,..., M) after division, the information indicating the approximate slant range change dAC ｍ after division, and the reception signal dS after divisionｍ (η) is obtained. The compensation amount calculation unit 13 calculates the actual slant range change dRC after division. ｍ and the change in the approximate slant range after splitting dAC ｍ The difference ΔC ｍ The compensation amount calculation unit 13 calculates the difference ΔC. ｍ Based on this, the received signal dS after division ｍ Compensation amount CA to compensate for the phase of (η) ｍ The compensation amount calculation unit 13 calculates the compensation amount CA. ｍ Information indicating this is output to the compensation unit 14.

[0014] The compensation unit 14 is implemented, for example, by the compensation circuit 34 shown in Figure 2. The compensation unit 14 receives the divided received signal dS from the division unit 12. ｍ (η) (m=1, ..., M) is obtained, and the compensation amount CA is calculated from the compensation amount calculation unit 13. ｍ Information indicating the amount CA is obtained. The compensation unit 14 obtains information indicating the amount of compensation CA ｍ Using this, the received signal dS ｍ The phase of (η) is compensated. The compensation unit 14 compensates for the received signal dS after phase compensation. ｍ (η)' is output to the range compression unit 15.

[0015] The range compression unit 15 is implemented, for example, by the range compression circuit 35 shown in Figure 2. The range compression unit 15 receives the phase-compensated received signal dS from the compensation unit 14. ｍ The (η)' (m=1, ..., M) is obtained. The range compression unit 15 receives the phase-compensated received signal dS ｍ (η)' is range-compressed. The range compression unit 15 compresses the range-compressed received signal RCS. ｍ (η) is output to the azimuth compression unit 18.

[0016] The time-distance calculation unit 16 is implemented, for example, by the time-distance calculation circuit 36 ​​shown in Figure 2. The time-distance calculation unit 16 obtains information indicating the approximation curve AC from the approximation curve calculation unit 11. Based on the approximation curve AC, the time-distance calculation unit 16 calculates the closest approach time η, which is the time when the platform 1 and the observation target Tgt are closest together. ０The closest approach distance R is the distance when platform 1 and the observation target Tgt are closest together. ０ The time distance calculation unit 16 calculates the closest approach time η. ０ Information indicating the closest approach distance R ０ Information indicating the distance R is output to the division unit 12 and the effective speed calculation unit 17, respectively. The time distance calculation unit 16 calculates the closest approach distance R ０ Information indicating this is output to the azimuth compression unit 18.

[0017] The effective speed calculation unit 17 is implemented, for example, by the effective speed calculation circuit 37 shown in Figure 2. The effective speed calculation unit 17 calculates the actual slant range change dRC after division from the division unit 12. ｍ Information indicating (m=1, ..., M) and the approximate slant range change dAC after division. ｍ The effective speed calculation unit 17 obtains information indicating the time of closest approach η from the time-distance calculation unit 16. ０ Information indicating the closest approach distance R ０ Information indicating the effective speed is obtained. The effective speed calculation unit 17 calculates the actual slant range change dRC after division. ｍ and the change in the approximate slant range after splitting dAC ｍ and the time of closest approach η ０ Closest approach distance R ０ Using the time zone TZ ｍ Effective speed Vr of platform 1 ｍ The system calculates the speed of platform 1 assuming that platform 1, which is moving on a non-linear track, is moving on a linear track. The effective speed calculation unit 17 calculates the speed of platform 1 in the time zone TZ ｍ Effective speed Vr of platform 1 ｍ Information indicating this is output to the azimuth compression unit 18.

[0018] The azimuth compression unit 18 is implemented, for example, by the azimuth compression circuit 38 shown in Figure 2. The azimuth compression unit 18 receives the range-compressed RCS signal from the range compression unit 15. ｍ The (η) (m=1, ..., M) is obtained. The azimuth compression unit 18 obtains the closest approach distance R from the time distance calculation unit 16. ０ Information indicating the time zone TZ is obtained, and the effective speed calculation unit 17 calculates the time zone TZ.ｍ Effective speed Vr of platform 1 ｍ Information indicating the effective speed Vr is obtained. The azimuth compression unit 18 is obtained. ｍ Closest approach distance R ０ Using this, the range-compressed received signal RCS ｍ (η) is azimuth compressed. The azimuth compression unit 18 compresses the received signal ACS after azimuth compression. ｍ (η) is output to the segmented image synthesis unit 19.

[0019] The segmented image synthesis unit 19 is implemented, for example, by the segmented image synthesis circuit 39 shown in Figure 2. The segmented image synthesis unit 19 receives the azimuth-compressed received signal ACS from the azimuth compression unit 18. ｍ (η) (m=1, ..., M) is obtained. The segmented image synthesis unit 19 receives the received signal ACS ｍ Multiple images reproduced based on (η) are combined. The segmented image combining unit 19 displays the combined image on a display device (not shown), for example.

[0020] In Figure 1, it is assumed that each of the components of the signal processing device 3—the approximation curve calculation unit 11, the division unit 12, the compensation amount calculation unit 13, the compensation unit 14, the range compression unit 15, the time distance calculation unit 16, the effective speed calculation unit 17, the azimuth compression unit 18, and the divided image synthesis unit 19—is implemented by dedicated hardware as shown in Figure 2. That is, it is assumed that the signal processing device 3 is implemented by the approximation curve calculation circuit 31, the division circuit 32, the compensation amount calculation circuit 33, the compensation circuit 34, the range compression circuit 35, the time distance calculation circuit 36, the effective speed calculation circuit 37, the azimuth compression circuit 38, and the divided image synthesis circuit 39. Each of the approximation curve calculation circuit 31, division circuit 32, compensation amount calculation circuit 33, compensation circuit 34, range compression circuit 35, time distance calculation circuit 36, effective speed calculation circuit 37, azimuth compression circuit 38, and division image synthesis circuit 39 can be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0021] The components of the signal processing device 3 are not limited to those implemented by dedicated hardware; the signal processing device 3 may also be implemented by software, firmware, or a combination of software and firmware. The software or firmware is stored as a program in the computer's memory. The computer refers to the hardware that executes the program, and includes, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a processor, or a DSP (Digital Signal Processor).

[0022] Figure 3 is a hardware configuration diagram of a computer when the signal processing device 3 is implemented by software or firmware. When the signal processing device 3 is implemented by software or firmware, programs that cause the computer to execute the respective processing procedures in the approximation curve calculation unit 11, the division unit 12, the compensation amount calculation unit 13, the compensation unit 14, the range compression unit 15, the time distance calculation unit 16, the effective speed calculation unit 17, the azimuth compression unit 18, and the divided image synthesis unit 19 are stored in memory 51. The computer's processor 52 then executes the programs stored in memory 51.

[0023] Furthermore, Figure 2 shows an example in which each component of the signal processing device 3 is implemented by dedicated hardware, and Figure 3 shows an example in which the signal processing device 3 is implemented by software or firmware, etc. However, this is only one example, and some components of the signal processing device 3 may be implemented by dedicated hardware, while the remaining components may be implemented by software or firmware, etc.

[0024] Next, the operation of the signal processing device 3 shown in Figure 1 will be explained. Figure 4 is a flowchart showing the signal processing method, which is the processing procedure of the signal processing device 3. If platform 1 is moving in a straight line at a constant velocity, and the trajectory of platform 1 is a straight trajectory as shown in Figure 5A, then the slant range R ｓｌ (η) is expressed as shown in equation (1) below. Slant range R over time ｓｌ The actual slant range change RC, which is a curve showing the change in (η), is a hyperbola, as shown in Figure 6. Figure 5A is an explanatory diagram showing the observation geometry of the region including the observed target Tgt when platform 1 is moving along a straight trajectory.

[0025] In equation (1), R ０ η is the distance when platform 1 and the observed object Tgt are closest (hereinafter referred to as the "closest approach distance"), Vs is the velocity of platform 1, and η ０ This is the time when platform 1 and the observation target Tgt are closest together (hereinafter referred to as the "closest approach time").

[0026] Platform 1 may be affected by, for example, non-constant outside air currents, in which case its orbit will deviate from a straight orbit and become a non-linear orbit. Alternatively, if Platform 1 is an artificial satellite orbiting the Earth, its orbit may become a non-linear orbit as shown in Figure 5B. For example, the slant range R when Platform 1's orbit is a non-linear orbit. ｓｌ (η) is R in equation (2) below. ｓｌｆｉｔ It is approximated as (η).

[0027] In equation (2), Vr is the effective speed of platform 1. The effective speed Vr of platform 1 is the hypothetical speed of platform 1 if it were assumed that platform 1, which is moving on a non-linear track, were moving on a linear track.

[0028] The radar 2 mounted on platform 1 emits radio waves toward the region containing the observation target Tgt, and then receives the radio waves after they have been reflected by the region containing the observation target Tgt. The radar 2 outputs the received radio wave signal S(η) to the splitting unit 12 of the signal processing device 3.

[0029] The approximation curve calculation unit 11 of the signal processing device 3 calculates the slant range R by performing a vector operation using the position of platform 1 at azimuth time η and the position of the observed object Tgt. ｓｌ (η) is calculated. The position of platform 1 at azimuth time η is known in the signal processing device 3, and the position of the observed object Tgt at azimuth time η is also known in the signal processing device 3. The approximation curve calculation unit 11 calculates the slant range R at multiple azimuth times η. ｓｌ By connecting (η), the actual slant range change RC can be determined.

[0030] Next, the approximation curve calculation unit 11 calculates the slant range R ｓｌ The coefficient C when approximating (η) by the following equation (3) ０ , C １ , C ２ This can be determined, for example, by using the least squares method.

[0031]

[0032] Next, the approximation curve calculation unit 11 calculates the coefficient C ０ , C １ , C ２ The determined slant range R of equation (3) ｓｌ (η) is the slant range R when the track of platform 1 is a straight track. ｓｌｆｉｔ As an approximation of (η), from equation (3), the slant range R at multiple azimuth times η is given by ｓｌｆｉｔ (η), that is, the approximate curve AC shown in Figure 6 is calculated (step ST1 in Figure 4). The approximate curve calculation unit 11 outputs information indicating the approximate curve AC to the time distance calculation unit 16.

[0033] FIG. 6 is an explanatory diagram showing an example of each of the approximate curve AC and the actual slant range change RC. In FIG. 6, the solid line indicates the approximate curve AC, and the dashed line indicates the actual slant range change RC. The horizontal axis is the azimuth time η, and the vertical axis is the slant range.

[0034] The time-distance calculation unit 16 acquires information indicating the approximate curve AC from the approximate curve calculation unit 11. Based on the approximate curve AC, the time-distance calculation unit 16 ０ calculates the closest time η ０ and the closest distance R ０ (step ST2 in FIG. 4). Specifically, as the closest distance R ｓｌｆｉｔ , the time-distance calculation unit 16 identifies the minimum slant range among the slant ranges R ０ (η) indicated by the approximate curve AC for all azimuth times η. The time-distance calculation unit 16 identifies the azimuth time η ０ corresponding to the minimum slant range R ０ as the closest time η. The time-distance calculation unit 16 outputs information indicating the closest time η ０ and information indicating the closest distance R ０ to the division unit 12 and the effective speed calculation unit 17, respectively. Further, the time-distance calculation unit 16 outputs information indicating the closest distance R

[0035] The division unit 12 acquires the received signal S(η) of the radio wave from the radar 2. The division unit 12 acquires information indicating the closest time η ０ and information indicating the closest distance R ０ from the time-distance calculation unit 16. The division unit 12 divides each of the actual slant range change RC and the received signal S(η) into M different time zones TZ １ to TZ Ｍis divided (step ST3 in FIG. 4). Information indicating the actual slant range change RC may be stored in the internal memory of the dividing unit 12, or may be given from outside the signal processing apparatus 3. Alternatively, the dividing unit 12 may calculate the actual slant range change RC as described above from the position of the platform 1 and the position of the observation target Tgt at the azimuth time η. The M time zones TZ １ ~TZ Ｍ each correspond to the observation time of each sub aperture in the multi-look processing shown in FIG. 8 described later. Therefore, if the number of sub apertures is M, each of the actual slant range change RC and the received signal S(η) is divided into M. As a result, the divided image synthesizing unit 19 described later synthesizes M images as images in units of sub apertures. The dividing unit 12 uses the closest time η ０ and the closest distance R ０ to approximate the actual slant range change dRC ｍ after division, thereby calculating an approximate slant range change dAC ｍ after division (step ST4 in FIG. 4).

[0036] FIG. 7 is an explanatory diagram showing an example of each of the actual slant range change dRC ｍ after division and the approximate slant range change dAC ｍ after division when M = 2. In FIG. 7, dAC ｍ (m = 1, 2) is the approximate slant range change after division, and dRC ｍ (m = 1, 2) is the actual slant range change after division. The horizontal axis is the azimuth time η, and the vertical axis is the slant range. The dividing unit 12 outputs information indicating the actual slant range change dRC ｍ after division, information indicating the approximate slant range change dAC ｍ after division, and the received signal dS ｍ (η) after division to the compensation amount calculation unit 13. The received signal dS ｍ (η) after division is the received signal in the time zone TZ ｍ corresponding to the approximate curve dAC ｍ after division. Further, the dividing unit 12 has the actual slant range change dRC after divisionｍ Information showing the change in approximate slant range after division, and dAC ｍ The splitting unit 12 outputs information indicating the effective speed to the effective speed calculation unit 17. Furthermore, the splitting unit 12 outputs the received signal dS after splitting. ｍ (η) is output to the compensation unit 14.

[0037] The effective speed calculation unit 17 calculates the actual slant range change dRC after division from the division unit 12. ｍ Information indicating (m=1, ..., M) and the approximate slant range change dAC after division. ｍ The effective speed calculation unit 17 obtains information indicating the time of closest approach η from the time-distance calculation unit 16. ０ Information indicating the closest approach distance R ０ Information indicating the effective speed is obtained. The effective speed calculation unit 17 calculates the actual slant range change dRC after division. ｍ and the change in the approximate slant range after splitting dAC ｍ and the time of closest approach η ０ Closest approach distance R ０ Using the time zone TZ ｍ Effective speed Vr of platform 1 ｍ Assuming that platform 1, which is moving on a non-linear track, is moving on a linear track, the speed of platform 1 is calculated (step ST5 in Figure 4). The effective speed calculation unit 17 calculates the time zone TZ ｍ Effective speed Vr of platform 1 ｍ Information indicating this is output to the azimuth compression unit 18.

[0038] The following is the effective speed Vr calculated by the effective speed calculation unit 17. ｍ The calculation process will be explained in detail. Time zone TZ after division ｍ In (m = 1, ..., M), the effective speed calculation unit 17 calculates the actual slant range change dRC after division as shown in the following equation (4). ｍ Slant Range R ｓｌｍ Calculate (η). Approximate curve dAC after division. ｍ Slant Range R ｓｌｆｉｔｍ (η) is the actual slant range change dRC after splitting. ｍ Slant Range R ｓｌｍAssuming that (η) is an approximation, the effective speed calculation unit 17 calculates the slant range R as shown in the following equation (5). ｓｌｍ (η) and the time of closest approach η ０ Closest approach distance R ０ Using the time zone TZ ｍ Effective speed Vr of platform 1 ｍ Calculate.

[0039]

[0040] The compensation amount calculation unit 13 calculates the actual slant range change dRC after division from the division unit 12. ｍ Information indicating (m=1, ..., M) and the approximate slant range change dAC after division. ｍ Information indicating the divided received signal dS ｍ (η) is obtained. The compensation amount calculation unit 13 calculates the actual slant range change dRC after division. ｍ and the change in the approximate slant range after splitting dAC ｍ The difference ΔC ｍ The compensation amount calculation unit 13 calculates the actual slant range change dRC after division, as shown in the following equation (6). ｍ Slant range R of azimuth time η in ｓｌ (η) and the change in the approximate slant range after splitting dAC ｍ Slant range R of azimuth time η in ｓｌｆｉｔ Difference ΔC from (η) ｍ Calculate (η). ΔC ｍ (η) = R ｓｌ (η)-R ｓｌｆｉｔ (η) (6)

[0041] The compensation amount calculation unit 13 calculates the difference ΔC as shown in the following equation (7): ｍ Based on (η), the received signal dS after division. ｍ Compensation amount CA to compensate for the phase of (η) ｍ Calculate (η) (step ST6 in Figure 4). Compensation amount CA ｍ (η) is the difference ΔC ｍ (η) is obtained by converting the phase. The compensation amount calculation unit 13 calculates the compensation amount CA ｍ Information indicating (η) is output to the compensation unit 14.

[0042] In equation (7), C is the speed of light, f ０ This is the central frequency of the radio waves emitted from radar 2.

[0043] The compensation unit 14 receives the divided received signal dS from the division unit 12. ｍ (η) (m=1, ..., M) is obtained, and the compensation amount CA is calculated from the compensation amount calculation unit 13. ｍ Information indicating (η) is obtained. The compensation unit 14 calculates the compensation amount CA as shown in the following equation (8). ｍ Using (η), the received signal dS ｍ The phase of (η) is compensated (step ST7 in Figure 4). The compensation unit 14 receives the received signal dS after phase compensation. ｍ (η)' is output to the range compression unit 15. dS ｍ (η)' = dS ｍ (η)×exp{-jCA ｍ (η)} (8)

[0044] The range compression unit 15 receives the phase-compensated received signal dS from the compensation unit 14. ｍ The (η)' (m=1, ..., M) is obtained. The range compression unit 15 receives the phase-compensated received signal dS ｍ (η)' is range-compressed (step ST8 in Figure 4). The received signal dS by the range-compression unit 15. ｍ The range compression process of (η)' itself is a well-known technique, so a detailed explanation will be omitted. The range compression unit 15 compresses the received signal RCS after range compression. ｍ (η) is output to the azimuth compression unit 18.

[0045] The azimuth compression unit 18 receives the range-compressed RCS signal from the range compression unit 15. ｍ (η) is obtained. The azimuth compression unit 18 obtains the closest approach distance R from the time distance calculation unit 16. ０ Information indicating the time zone TZ is obtained, and the effective speed calculation unit 17 calculates the time zone TZ. ｍ Effective speed Vr of platform 1 ｍ Information indicating the effective speed Vr is obtained. The azimuth compression unit 18 is obtained. ｍ Closest approach distance R ０Using this, the range-compressed received signal RCS ｍ (η) is azimuth compressed (step ST9 in Figure 4). The azimuth compression unit 18 compresses the received signal ACS after azimuth compression. ｍ (η) is output to the segmented image synthesis unit 19.

[0046] The following is the received signal RCS by the azimuth compression unit 18. ｍ The azimuth compression process of (η) will be explained in detail. The azimuth compression unit 18 operates at an effective speed Vr as shown in the following equation (9). ｍ Closest approach distance R ０ Using and the azimuth reference function H ａｚ (f η ) calculates the azimuth reference function H ａｚ (f η ) is defined in the Doppler frequency domain, which is the azimuth frequency domain.

[0047] In equation (9), f η This is the azimuth frequency.

[0048] The azimuth compression unit 18 compresses the received signal RCS after range compression. ｍ (η) is Fourier transformed in the azimuth direction. The azimuth compression unit 18 performs the Fourier transform on the received signal RCS as shown in equation (10) below. ｍ (f η ) and the azimuth reference function H ａｚ (f η Multiply by ). MR ｍ (f η ) = RCS ｍ (f η ) × H ａｚ (f η ) (10) The azimuth compression unit 18 receives the RCS signal ｍ (f η ) and the azimuth reference function H ａｚ (f η MR is the result of multiplication with ) ｍ (f η The AZI-compression unit 18 performs an inverse Fourier transform on the AZI-compressed received signal ACS ｍ (η) as MR ｍ (f ηThe inverse Fourier transform result of ) is output to the segmented image synthesis unit 19. Note that the azimuth compression function is not limited to this one; any function with equivalent functionality will suffice.

[0049] The segmented image synthesis unit 19 receives M azimuth-compressed received signals ACS from the azimuth compression unit 18. １ (η) ~ ACS Ｍ (η) is obtained. The segmented image synthesis unit 19 receives the received signal ACS ｍ M images are synthesized based on (η) (m=1, ..., M) (step ST10 in Figure 4). Received signal ACS ｍ The image synthesis process based on (η) is a well-known technique, so a detailed explanation is omitted. The segmented image synthesis unit 19 displays the synthesized image on a display device (not shown), for example.

[0050] The image synthesis process performed by the divided image synthesis unit 19, which combines M images, corresponds to the multi-look processing shown in Figure 8. Figure 8 is an explanatory diagram illustrating multi-look processing. Multi-look processing is a process that divides the composite aperture into several independent sub-apertures (labeled "looks" in Figure 8), reconstructs the image in look units, and synthesizes multiple images. In multi-look processing with M looks, the noise dispersion is reduced to 1 / √M. As a result, fluctuations in intensity present in the synthesized image are suppressed, improving visibility. Note that in order to achieve multi-look processing with M looks while maintaining resolution, it is necessary to increase the observation time by M times and the Doppler bandwidth by M times compared to normal observation.

[0051] In the above embodiment 1, the signal processing device 3 is configured to include an approximation curve calculation unit 11 that calculates an approximation curve of the actual slant range change, which is a curve showing the actual distance change over time between the platform 1 equipped with a radar that emits radio waves toward the object of observation and receives the radio waves after they have been reflected by the object of observation, and the object of observation; and a division unit 12 that divides the actual slant range change and the received signal of the radio waves by the radar into multiple different time periods and calculates the respective approximate slant range changes by approximating the actual slant range changes after division. The signal processing device 3 is also configured to include a compensation amount calculation unit 13 that calculates a compensation amount to compensate for the phase of each received signal after division by the division unit 12 based on the difference between each actual slant range change after division by the division unit 12 and each approximate slant range change calculated by the division unit 12; a compensation unit 14 that compensates for the phase of each received signal after division by the division unit 12 using the respective compensation amounts calculated by the compensation amount calculation unit 13; and a divided image synthesis unit 19 that synthesizes multiple images reconstructed based on each received signal after phase compensation by the compensation unit 14. Therefore, the signal processing device 3 can suppress the degradation of image reproduction accuracy even in situations where approximating the actual slant range change using a single curve, such as with long-term observation data, would result in a large approximation error.

[0052] In Embodiment 1, the signal processing device 3 is configured such that a time-distance calculation unit 16 calculates the closest approach time, which is the time when the platform 1 and the object of observation are closest, and the closest approach distance, which is the distance when the platform 1 and the object of observation are closest, based on the approximate curve calculated by the approximate curve calculation unit 11; an effective speed calculation unit 17 calculates the effective speed of the platform 1 in each time period, assuming that the platform 1, which is moving on a non-linear trajectory, is moving on a linear trajectory, using the approximate slant range change, closest approach time, and closest approach distance calculated by the division unit 12; and an azimuth compression unit 18 azimuth compresses each received signal after phase compensation by the compensation unit 14, using the effective speed and closest approach distance for each time period calculated by the effective speed calculation unit 17; and a divided image synthesis unit 19 synthesizes a plurality of images reconstructed based on each received signal after azimuth compression by the azimuth compression unit 18. Therefore, the signal processing device 3 can suppress the degradation of image reproduction accuracy even when the error of the approximation curve for the actual slant range change is different, and can also improve the resolution in the direction of movement of the platform 1.

[0053] In Embodiment 1, the signal processing device 3 is configured such that it includes a range compression unit 15 that range-compresses each received signal after phase compensation by the compensation unit 14, and an azimuth compression unit 18 that uses the effective speed and closest approach distance for each time period calculated by the effective speed calculation unit 17 to azimuth-compress each received signal after range compression by the range compression unit 15. Therefore, the signal processing device 3 can improve the resolution of the direction of movement of the platform 1 and the direction perpendicular to the direction of movement of the platform 1.

[0054] In Embodiment 1, the signal processing device 3 is configured to divide the image into M independent sub-apertures using a division unit 12. Therefore, the signal processing device 3 performs a synthesis process in the divided image synthesis unit 19 to add the M images noncoherently, which can suppress fluctuations in intensity on the image compared to the image before synthesis, thereby improving visibility. Furthermore, by coherently adding the M images while retaining their phase information, it is also possible to generate an image with higher resolution compared to the image before synthesis.

[0055] It should be noted that this disclosure allows for modifications of any component of the embodiment, or the omission of any component of the embodiment.

[0056] The signal processing device of this disclosure can be used as a synthetic aperture radar.

[0057] 1 Platform, 2 Radar, 3 Signal Processing Unit, 11 Approximation Curve Calculation Unit, 12 Segmentation Unit, 13 Compensation Amount Calculation Unit, 14 Compensation Unit, 15 Range Compression Unit, 16 Time Distance Calculation Unit, 17 Effective Speed ​​Calculation Unit, 18 Azimuth Compression Unit, 19 Segmented Image Synthesis Unit, 31 Approximation Curve Calculation Circuit, 32 Segmentation Circuit, 33 Compensation Amount Calculation Circuit, 34 Compensation Circuit, 35 Range Compression Circuit, 36 Time Distance Calculation Circuit, 37 Effective Speed ​​Calculation Circuit, 38 Azimuth Compression Circuit, 39 Segmented Image Synthesis Circuit, 51 Memory, 52 Processor.

Claims

1. A signal processing device comprising: an approximation curve calculation unit that calculates an approximation curve of the actual slant range change, which is a curve showing the actual distance change over time between a platform equipped with a radar that emits radio waves toward an observation target and receives the radio waves after they have been reflected by the observation target; a division unit that divides the actual slant range change and the received signal of the radio waves from the radar into a plurality of mutually different time periods and calculates the respective approximate slant range change by approximating the actual slant range change after each division; a compensation amount calculation unit that calculates a compensation amount to compensate for the phase of each received signal after division by the division unit based on the difference between each actual slant range change after division by the division unit and each approximate slant range change calculated by the division unit; a compensation unit that compensates for the phase of each received signal after division by the division unit using the respective compensation amounts calculated by the compensation amount calculation unit; and a divided image synthesis unit that synthesizes a plurality of images reconstructed based on each received signal after phase compensation by the compensation unit.

2. The signal processing apparatus according to claim 1, characterized in that the approximation curve calculation unit calculates the slant range, which is the distance between the platform and the observation target at multiple times when the platform is moving on a non-linear track, and calculates the approximation curve from the slant range at multiple times.

3. A signal processing apparatus according to claim 1 or 2, comprising: a time-distance calculation unit that calculates the closest approach time, which is the time when the platform and the object of observation are closest, and the closest approach distance, which is the distance when the platform and the object of observation are closest, based on the approximate curve calculated by the approximate curve calculation unit; an effective speed calculation unit that calculates the effective speed of the platform in each time period, assuming that the platform moving on a non-linear trajectory is moving on a linear trajectory, using the approximate slant range change calculated by the division unit, the closest approach time, and the closest approach distance; and an azimuth compression unit that azimuth compresses each received signal after phase compensation by the compensation unit, using the effective speed in each time period calculated by the effective speed calculation unit and the closest approach distance, wherein the division image synthesis unit synthesizes a plurality of images reconstructed based on each received signal after azimuth compression by the azimuth compression unit.

4. The signal processing apparatus according to claim 3, further comprising a range compression unit for range-compressing each received signal after phase compensation by the compensation unit, wherein the azimuth compression unit azimuth-compresses each received signal after range compression by the range compression unit using the effective speed in each time period calculated by the effective speed calculation unit and the closest approach distance.

5. The signal processing apparatus according to any one of claims 1 to 4, characterized in that the durations of the multiple time periods related to the division section are different from each other.

6. A signal processing method comprising: an approximation curve calculation unit calculates an approximation curve of the actual slant range change, which is a curve showing the actual distance change over time between a platform equipped with a radar that receives radio waves reflected by the observation target after radiating radio waves toward the observation target and the observation target; a division unit divides the actual slant range change and the received signal of the radio waves by the radar into multiple different time periods and calculates each approximated slant range change by approximating each actual slant range change after division; a compensation amount calculation unit calculates a compensation amount to compensate for the phase of each received signal after division by the division unit based on the difference between each actual slant range change after division by the division unit and each approximated slant range change calculated by the division unit; a compensation unit compensates for the phase of each received signal after division by the division unit using each compensation amount calculated by the compensation amount calculation unit; and a divided image synthesis unit synthesizes multiple images reconstructed based on each received signal after phase compensation by the compensation unit.

7. A signal processing system comprising: a platform equipped with a radar that emits radio waves toward an object to be observed and then receives the radio waves after they have been reflected by the object to be observed; an approximation curve calculation unit that calculates an approximation curve of the actual slant range change, which is a curve showing the actual change in distance between the platform and the object to be observed over time; a division unit that divides the actual slant range change and the received signal of the radio waves from the radar into a plurality of mutually different time periods and calculates the respective approximate slant range changes by approximating the actual slant range changes after division; a compensation amount calculation unit that calculates a compensation amount to compensate for the phase of each received signal after division by the division unit based on the difference between each actual slant range change after division by the division unit and each approximate slant range change calculated by the division unit; a compensation unit that compensates for the phase of each received signal after division by the division unit using the respective compensation amounts calculated by the compensation amount calculation unit; and a divided image synthesis unit that synthesizes a plurality of images reconstructed based on each received signal after phase compensation by the compensation unit.

Images