JP7413106B2 - Radar device and radar signal processing method - Google Patents

Description

The present embodiment relates to a radar device and a radar signal processing method.

In radar devices, it is necessary to suppress false detections due to clutter and detect targets with high precision from low speeds to high speeds. Furthermore, in the case of a target such as a small flying object with a blade coupled to the body, it is desired to detect parts such as the body and the blade when flying at low speed.

However, with conventional radar equipment, low-speed targets are greatly affected by clutter (including nearby reflected waves), so in environments with strong clutter, it becomes difficult to distinguish between clutter and the target, and if only the target area is used. First, there was the problem that the target itself could not be detected.

MTI, Yoshida, ‘Revised Radar Technology’, Institute of Electronics, Information and Communication Engineers, pp.67-77 (1996) FIR Filter, Mitani, Digital Filter Design, Shokodo, pp.82-86 (1987) CFAR (CA-CFAR, SO-CFAR, GO-CFAR), Sekine, Radar signal processing technology, Institute of Electronics, Information and Communication Engineers, pp.96-103 (1991) 2D CFAR, Guy Morris, Airborne Pulsed Doppler Radar 2nd edition, Artech House, pp.399-417 (1996) Correlation tracking, Yoshida, ‘Revised radar technology’, Institute of Electronics, Information and Communication Engineers, pp.254-259 (1996)

As mentioned above, conventional radar equipment has a problem in that it is unable to eliminate the influence of clutter when the target is at a low speed in an environment of strong clutter, and is unable to detect the target or the part of the target. .

An object of the present embodiment is to provide a radar device and a radar signal processing method that can accurately detect a target or a target part from low speed to high speed even in a strong clutter environment.

In order to solve the above problems, the radar device according to the present embodiment extracts a CPI (Coherent Pulse Interval) signal from a received signal of a transmitted pulse, outputs a range-Doppler range from the CPI signal, and outputs a range-Doppler range from the CPI signal. Preprocessing is performed to suppress clutter components within the Doppler range, CFAR (Constant False Alarm Rate) processing is performed to extract Doppler cell reflection points from the range where the target is present, and a first step is performed to discriminate between the target and clutter from the output of the CFAR processing. 1 discrimination processing is performed, an observed value by CPI is output from the discrimination result of the target and clutter, and the observed value is subjected to correlation processing and tracking processing at a predetermined period for the entire space within the search range, and the correlation processing and tracking processing are performed. A second discrimination process is performed to suppress false detections from the output, discriminate between clutter and a target, and output a target signal. In the first region, a one-dimensional or two-dimensional first type CFAR is applied; in the second region, a one-dimensional or two-dimensional second type CFAR is applied; 1 discrimination processing performs discrimination processing for clutter and a target on the detection value of a third region that is an integration of the first region and the second region, and the second discrimination processing section performs the discrimination processing for clutter and the target. The results are used to identify the target body or blade.

FIG. 2 is a block diagram showing the configuration of a transmitting section, a receiving section, and a signal processing section of a radar device according to an embodiment. 2 is a flowchart showing the processing contents of the signal processing section in FIG. 1. FIG. 2 is a flowchart showing the flow of preprocessing 1 of the signal processing unit in FIG. 1; FIG. 4 is a conceptual diagram showing how clutter is suppressed by preprocessing 1 in FIG. 3; 2 is a flowchart showing the process flow of preprocessing 2 of the signal processing unit in FIG. 1. FIG. FIG. 6 is a diagram showing the range-Doppler range extracted in preprocessing 2 of FIG. 5; FIG. 2 is a diagram illustrating an example of processing contents of pre-processing 2 in FIG. 1; FIG. 2 is a diagram showing another example of processing contents of preprocessing 2 in FIG. 1; FIG. 2 is a diagram defining a region A around 0 Doppler and a region B outside around 0 Doppler in the target peripheral region of the range-Doppler axis when applying CFAR in FIG. 1; FIG. 2 is a diagram showing CA-CFAR applied to the CFAR processing in FIG. 1. FIG. 2 is a diagram showing GO-CFAR applied to the CFAR processing in FIG. 1. FIG. 2 is a diagram showing SO-CFAR applied to the CFAR processing in FIG. 1. FIG. 2 is a diagram showing two-dimensional CFAR of the range-Doppler axis applied to the CFAR processing of FIG. 1. 2 is a flowchart showing the processing contents of the first embodiment for extracting a detection prohibited range in the first discrimination processing of FIG. 1; FIG. 15 is a diagram showing the state of the first discrimination processing in the first embodiment of FIG. 14; 2 is a flowchart showing the processing contents of a second embodiment for extracting a detection prohibited range in the first discrimination processing of FIG. 1; FIG. 17 is a diagram showing the state of the first discrimination processing in the second embodiment of FIG. 16; 2 is a flowchart showing the processing contents of a third embodiment for extracting a detection prohibited range in the first discrimination processing of FIG. 1; FIG. 19 is a diagram showing the state of the first discrimination processing in the third embodiment of FIG. 18; 3 is a flowchart showing the processing contents of the fourth embodiment for discriminating blade signals in the first discrimination processing of FIG. 1; FIG. 21 is a diagram showing the state of the first discrimination processing in the fourth embodiment of FIG. 20; 2 is a flowchart showing the processing contents of the second discrimination processing of FIG. 1;

Embodiments will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a transmitting section, a receiving section, and a signal processing section of a radar device according to an embodiment. In FIG. 1, in the transmission section, a transmission signal generator 11 generates a transmission signal using PRI (Pulse Repetition Interval) pulses, a DA converter 12 converts it into an analog signal, and a frequency converter 13 converts it into a radio frequency (RF) signal. The signal is converted, power amplified by a high-output amplifier 14, and transmitted from an antenna 16 via a circulator 15.

In the receiving section, a reflected signal from a target etc. is captured by an antenna 16, sent and received by a circulator 15, amplified by reducing noise by a low noise amplifier 17, and frequency converted to baseband by a frequency converter 18. The AD converter 19 converts it into a digital signal.

In the signal processing unit, the PCI signal extractor 21 extracts a CPI signal from the radar reception signal (PRI signal) of N (N≧1) hit transmission pulses (hereinafter referred to as N pulses) output from the AD converter 19. , the range-Doppler output device 22 obtains two-dimensional data of fast-time and slow-time between PRIs for each range cell from the CPI signal. Next, a preprocessor 23 suppresses the clutter component, a CFER (Constant False Alarm Rate) 24 extracts the range-Doppler cell reflection point where the target exists, and a first discrimination processor 25 extracts the range-Doppler cell reflection point where the target exists. The observed value output unit 26 outputs the observed value based on the CPI based on the discrimination result, and the correlation tracker 27 performs correlation processing and tracking processing at a predetermined period for the entire space within the search range, and performs a second discrimination process. A device 28 discriminates between the target body and the blade, and outputs a low-speed target signal.

The processing contents of the signal processing section will be explained along the processing flow shown in FIG. 2.
In the radar device, when the target detection process is started, the CPI signal extracted from the radar reception signal in the PCI signal extraction process (21) is input (step S11). This CPI signal is two-dimensional data of fast-time (range axis) for each range cell and slow-time (Doppler axis) between PRIs. This CPI signal includes a target signal and clutter components. The clutter component is a low frequency component near Doppler 0, especially in the case of ground clutter received by a ground fixed radar. On the other hand, unless the target is a fixed target, the target signal has a Doppler component due to velocity.

Next, in range-Doppler output processing (22), the RD (range-Doppler) signal is extracted from the CPI signal and output (step S12). FIG. 3 shows an example of the target signal and clutter component in the RD signal. In FIG. 3, the horizontal axis is the range (fast-time) axis, and the vertical axis is the Doppler (slow-time) axis, and the target signal has a signal component of the fuselage and a signal component of the blade extending in the direction of the Doppler axis. On the other hand, in addition to the main lobe of the Doppler axis, the clutter component becomes a signal in which side lobes (SL) spread in the Doppler axis direction if the intensity is strong.

Next, the clutter component is suppressed as much as possible by preprocessing (23) (step S13), and then the reflection point of the range-Doppler cell where the target exists is extracted by CFAR processing (24) (steps S14, S15, S16). This reflection point contains a clutter component and a target signal. Details of the CFAR processing will be described later. Next, the target signal and the clutter component are discriminated by the discrimination process (25) (step S17), and the observed value by CPI is output by the observed value output process (26) (step S18).

Next, the entire space within the search range is treated as one frame, and correlation processing and tracking processing (27) are performed as processing between frames in the same space (beams in the same direction) (see Non-Patent Document 5) (Step S19) . Here, the correlation process is a process that uses a gate to determine whether the observed value in the previous frame and the observed value in the current frame are the same target. Further, the tracking process is a tracking filter process that uses the predicted value of the previous cycle and the observed value of the current cycle to calculate a smoothed value and a predicted value of the next frame. Next, using the smoothed value (predicted value) of the correlation tracking process result, a process (28) is performed to discriminate between the clutter component and the target signal (step S20). As a result of the above processing, only the target signal can be output.

In the above processing, the features of this embodiment will be explained.
First, the preprocessing (23) before applying CFAR will be described (step S13 in FIG. 2). The preprocessing includes preprocessing 1 in which clutter is suppressed using a Doppler axis filter, and preprocessing 2 in which clutter is suppressed by flattening the clutter intensity.

Pre-processing 1 will be explained using FIGS. 3 and 4. FIG. 3 is a flowchart showing the process flow of preprocessing 1, and FIG. 4 is a conceptual diagram showing how clutter is suppressed by preprocessing 1.

In preprocessing 1, as shown in FIG. 3, when range-Doppler data is input (step S21), a predetermined threshold is set, and a range-Doppler range with a strong clutter intensity exceeding this threshold is extracted (step S22). ). Next, using the complex signal on the slow-time axis in the extraction range range, a frequency filter is applied that forms a null in the strong clutter range (0 Doppler) on the Doppler axis (step S23). As this frequency filter, an MTI (see Non-Patent Document 1), an FIR filter based on the frequency axis inverse sampling method (see Non-Patent Document 2), etc. can be applied. Finally, the range-Doppler data after 0 Doppler suppression is output (step S24). This situation is shown in FIG. Figure 4(a) shows how a null is formed in the frequency response characteristic, Figure 4(b) shows the extracted range-Doppler range that includes clutter intensity exceeding the threshold, and Figure 4(c) shows the state in which a null is formed in the frequency response characteristic. The clutter region shown in FIG. 4(b) has been removed by the null formation shown in FIG. 4(a), and the Doppler null range has been identified.

Next, preprocessing 2 will be explained using FIGS. 5 to 8. FIG. 5 is a flowchart showing the process flow of preprocessing 2, FIG. 6 is a diagram showing the range-Doppler range extracted in preprocessing 2, FIG. 7 is a diagram showing an example of the processing content of preprocessing 2, and FIG. 2 is a diagram illustrating another example of processing contents of pre-processing 2. FIG.

In preprocessing 2, as shown in FIG. 5, when range-Doppler data is input (step S31), a predetermined threshold is set, and a range-doppler range with a strong clutter intensity exceeding this threshold is extracted (step S32). ). Next, in the extracted range-Doppler range, a low-order approximate curve is calculated using the amplitude signal of the range (fast-time) axis for each cell of the Doppler axis (step S33). This can be formulated as follows. For the approximate curve, coefficients may be determined by the method of least squares or the like using, for example, a polynomial approximation expression shown in the following equation.

At this time, if the order M is increased, it approximates a target signal that is higher than the clutter component, so it is necessary to set it to a low order (for example, about 1 to 3) so as not to suppress the expected target signal. The original signal is divided using this approximate curve (step S33). This makes it possible to flatten the entire level. This situation is shown in FIG. Figure 7(a) shows how a low-order approximate curve is calculated from the amplitude signal of the range axis extracted for each cell of the range-Doppler axis, and Figure 7(b) shows the low-order approximate curve calculated in Figure 7(a). The following approximate curve is shown, and FIG. 7(c) shows the result of flattening the clutter component by dividing the original signal by the low-order approximate curve shown in FIG. 7(b).

In order to flatten the clutter described above, in addition to a low-order approximate curve, there is also a method of dividing by an approximate curve based on a moving average. This situation is shown in FIG. Figure 8(a) shows how the moving average value of the amplitude signal of the range axis extracted for each cell of the range-Doppler axis is calculated in units of L cells along the fast-time axis. Fig. 8(a) shows an approximate curve based on the moving average value calculated, and Fig. 8(c) shows the result of flattening the clutter component by dividing the original signal by the approximate curve shown in Fig. 8(b). ing. In this case, the moving average calculates an approximate curve using the amplitude average value of a range of L (L≧2) cells shifted by one cell on the fast-time axis. Thereby, the clutter level can be flattened by dividing the original signal.

Next, CFAR (see Non-Patent Document 3) is applied to the signal after the preprocessing has been applied (steps S14 to S16 in FIG. 2). In describing the subsequent processing, as shown in FIG. 9, regarding the target surrounding area on the range (fast-time) axis and the Doppler (slow-time) axis, area A near 0 Doppler and area B other than near 0 Doppler will be used. Define the area. CFAR includes one-dimensional CFAR with a range axis or Doppler axis shown in FIGS. 10 to 12, and two-dimensional CFAR with a range-Doppler axis shown in FIG. 13 (see Non-Patent Document 4). In two-dimensional CFAR, the number of reference cells increases compared to one-dimensional CFAR, so it is easier to obtain stable CFAR output and false detections can be reduced.

Various one-dimensional CFARs are shown in Figures 10 to 12. In FIGS. 10 to 12, CFAR is a process of shifting the test cells to be detected by one cell along the axis of symmetry and comparing them with a threshold to detect cells exceeding the threshold. A guard cell is set around this test cell, and a reference cell is set around it. CA-CFAR in FIG. 10 (see Non-Patent Document 3) shows the average amplitude value of the reference cell, GO-CFAR in FIG. 11 (see Non-Patent Document 3) shows the maximum amplitude value of the reference cell, In Patent Document 3), the CFAR output obtained by dividing each input signal using the minimum amplitude value of the reference cell is compared with a predetermined threshold, and when the threshold is exceeded, the detection bank is increased. The same applies to the two-dimensional CFAR shown in FIG.

Compared to CA-CFAR, GO-CFAR tends to reduce false positives, but the detection probability tends to decrease. On the other hand, SO-CFAR can detect close-packed targets, but the number of false detections increases. Here, the width of the guard cell between the test cell and the reference cell determines the width of the target that can be detected, and if the target width and the clutter width are different, discrimination is possible.

Based on the above, if the shapes of the clutter and the target are clearly different in the RD data near the main lobe (ML) of the Doppler axis, by changing the number of guard cells, normal CA-CFAR or GO-CFAR can be used. Although it is possible to detect and distinguish between clutter and a target, the shapes of the target and clutter in the main lobe ML of the Doppler axis are similar, and they cannot be distinguished using only the main lobe ML. Therefore, Doppler axis sidelobe (SL) information is required. To detect this using CFAR, CFAR detection is required to extract nearby reflection points. SO-CFAR is used for this detection.

As described above, CA-CFAR (GO-CFAR) is applied to area A, and SO-CFAR is applied to area B.

Furthermore, when using one-dimensional CFAR, it is necessary to consider the number of cells required for CFAR processing and the spread of reflection points when selecting the range axis or Doppler axis. When the clutter intensity is strong, the side lobe SL spreads on the Doppler axis in the clutter range, so it is desirable to select the Doppler axis to suppress clutter and detect the target. However, if the number of PRIs (number of pulse hits) is smaller than the number of reference cells required for CFAR processing, the Doppler axis cannot be selected, and the range axis will be selected. In this case, with strong clutter, many clutter sidelobes SL may be detected on the Doppler axis, so the number of detected cells on the Doppler axis for each range cell is counted, and the sum of the number of detected cells exceeds a predetermined threshold. If so, the range cell performs processing such as rejecting the detection as clutter (discrimination processing 28, step S20).

The first target and clutter discrimination processing (25, step S17) will be described below for the CFAR detection results of the A region and the B region.

In the first discrimination process, when the clutter intensity is strong, the side lobe SL of the Doppler axis is also high, and detection is prohibited in that range because false detection increases. FIG. 14 shows a processing flow of the first embodiment for extracting this range. In FIG. 14, the range-Doppler data before processing is input (step S41), and the maximum amplitude value of the Doppler axis is extracted for each range cell (step S42). For this maximum amplitude value, a predetermined amplitude threshold for the strong clutter range is set, a range exceeding this is extracted (step S43), the range exceeding the amplitude threshold is set as a detection prohibited range, and the rest can be detected. range (step S44).

FIG. 15 shows the state of the first discrimination process. Figure 15(a) shows strong clutter components (including ships, aircraft, etc.) and targets (target fuselage, blades) within the range-Doppler range, and Figure 15(b) shows the maximum amplitude of the Doppler axis for each range cell. The figure shows how the detection range is compared with the threshold, and the range exceeding the threshold is set as the detection prohibited range, and the rest is set as the detectable range.

Next, FIG. 16 shows a processing flow of a second embodiment for specifying the detection prohibited range. In FIG. 16, the detection results after CFAR processing are input (step S51), and the addition result of the number of detection cells of the Doppler axis is output for each range cell (step S52). For this addition result, a predetermined detection number threshold for the strong clutter range is set, range cells exceeding this are extracted (step S53), the range exceeding the threshold is set as a detection prohibited range, and the rest is set as a detectable range. (Step S54). This situation is shown in FIG. Figure 17(a) shows strong clutter components (including ships, aircraft, etc.) and targets (target fuselage, blades) within the range-Doppler range, and Figure 17(b) shows the number of Doppler axis detection cells for each range cell. The result of addition is compared with a threshold, and the range exceeding the threshold is set as a detection prohibited range, and the rest is set as a detectable range. Since it is necessary to extract the Doppler axis clutter SL and leave the blade, the predetermined threshold is determined by taking both into consideration.

Next, FIG. 18 shows a processing flow of a third embodiment for specifying the detection prohibited range. In the third embodiment, the CFAR detection results include false detections due to clutter, and when comparing these false detections with target detection, the characteristic is that the target often has multiple detection cells around the detection point. The difference is used to suppress false detection of clutter. In FIG. 18, the detection results after CFAR processing are input (step S61), a predetermined range Doppler cell range is set around each detection range cell, and the number of cells within the range is calculated (step S62). Cells whose number of cells exceeds a predetermined threshold are extracted (step S63). Furthermore, a predetermined range-Doppler range is set for each extracted cell, and the target and clutter cells within that range are each integrated into one (step S64). Thereby, the number of detection points can be reduced and the processing load of subsequent correlation processing and tracking processing can be reduced. This situation is shown in FIG. FIG. 19(a) shows the distribution of detection range cells after CFAR processing in the range-Doppler range and the predetermined range-Doppler cell range set around each detection range cell, and FIG. 19(b) shows the number of cells within the predetermined range. indicates cells exceeding the threshold, and FIG. 19(c) shows how target and clutter cells within a predetermined range are each integrated into one cell.

After performing various discrimination processes as in the first to third embodiments, if there is a signal due to the rotation of the blade (hereinafter referred to as a blade signal), it is necessary to extract it. This has the advantage that a predetermined target can be detected even if the target torso is suppressed by the frequency filter. Since the blade signal spreads along the Doppler axis near the same range cell, it can be extracted using this.

FIG. 20 shows a processing flow of a fourth embodiment for discriminating blade signals in the first discrimination process. In FIG. 20, the detection result after CFAR processing is input (step S71), and the addition result of the number of detection cells of the Doppler axis is output for each range cell (step S72). For this addition result, a predetermined detection number threshold for the blade signal is set, range cells exceeding this threshold are extracted (step S73), and the range range of the extracted cells is extracted as a blade signal (step S74). This situation is shown in FIG. FIG. 21(a) shows the clutter component and target (target body, blade) within the range-Doppler range, and FIG. 21(b) compares the addition result of the number of detection cells of the Doppler axis for each range cell and the threshold, This shows how the range exceeding the threshold is determined to be a blade, and the rest are removed.

As a result of the above processing, the observed value, which is a process within the CPI, can be output in the observed value output process (26) (step 18 in FIG. 2). Next, the entire space within the search range is treated as one frame, and correlation processing and tracking processing (27) are performed as processing between frames in the same space (beams in the same direction) (step 19 in FIG. 2). The correlation process is a process that uses a gate to determine whether the observed value in the previous frame and the simple observed value in the current frame are the same target. The tracking process is a filtering process that uses the predicted value of the previous cycle and the observed value of the current cycle to calculate a smoothed value and a predicted value of the next frame. Using this smoothed value (predicted value), a second discrimination process (28, step S20) is performed to discriminate between clutter and the target.

FIG. 22 shows the processing flow of the second discrimination process. In FIG. 22, a smoothed value (predicted value) speed is extracted from the observed value (step S81), and the blade signal is set as a target (step S82). Here, cells whose speeds extracted in step S81 exceed a predetermined threshold are extracted as targets (step S83). The target extracted in step S83 and the target obtained in step S82 are output as extracted targets (step S84). That is, the observed values include the blade signal extracted in the first discrimination process and others. Regarding the blade signal, a predetermined target is set. For other cases, the speed of the smoothed value (predicted value) is used.Since the clutter has a speed near 0 and the target has a speed other than that, the target signal can be extracted by setting a predetermined speed threshold. I can do it. As a result, a target can be extracted based on the blade signal even when the torso speed is 0, in addition to when the torso speed is other than 0.

As described above, the radar device according to the above embodiment performs CFAR preprocessing on the range-Doppler axis signal (RD signal), and then divides the range-Doppler axis signal (RD signal) into area A near Doppler and area B other than the area. , one-dimensional or two-dimensional CA-CFAR is applied in area A, one-dimensional or two-dimensional SO-CFAR is applied in area B, and for the detected value of area AB, which is the integration of area A and B, Discrimination processing is performed on clutter and targets, and the results of correlation and tracking using signal-processed observed values are used to reduce false detections, suppressing clutter and detecting target torso or blade signals with high precision. can be detected.

It should be noted that the present invention is not limited to the above-described embodiments as they are, but can be embodied by modifying the constituent elements within the scope of the invention at the implementation stage. Moreover, various inventions can be formed by appropriately combining the plurality of components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiments. Furthermore, components from different embodiments may be combined as appropriate.

11... Transmission signal generator, 12... DA converter, 13... Frequency converter, 14... High output amplifier, 15... Circulator, 16... Antenna, 17... Low noise amplifier, 18... Frequency converter, 19... AD converter , 21...PCI signal extractor, 22...range-Doppler output device, 23...preprocessor, 24...CFER, 25...first discrimination processor, 26...observation value output device, 27...correlation tracker, 28...th 2 discrimination processor.

Claims (10)

a CPI signal extraction unit that extracts a CPI (Coherent Pulse Interval) signal from the received signal of the transmitted pulse;
a range-Doppler output section that outputs a range-Doppler range from the CPI signal;
a CFAR processing unit that performs preprocessing to suppress clutter components within the range-Doppler range and performs CFAR (Constant False Alarm Rate) processing to extract a range-Doppler cell reflection point where a target exists;
a first discrimination processing unit that discriminates between a target and clutter from the output of the CFAR processing unit;
an observed value output unit that outputs an observed value by CPI from the target and clutter discrimination results;
a correlation tracking unit that performs correlation processing and tracking processing on the observed values at a predetermined period for the entire space within the search range;
a second discrimination processing unit that suppresses false detection from the outputs of the correlation processing and tracking processing, discriminates between clutter and a target, and outputs a target signal;
After performing the preprocessing, the CFAR processing unit divides the region into a first region near 0 Doppler and a second region other than that, and applies one-dimensional or two-dimensional first type CFAR to the first region. However, in the second region, a one-dimensional or two-dimensional second type CFAR is applied,
The first discrimination processing unit performs discrimination processing for clutter and a target on the detection value of a third region that is an integration of the first region and the second region,
The second discrimination processing unit is a radar device that discriminates a target body or blade using the results of the correlation processing and tracking processing. 2. The radar apparatus according to claim 1, wherein the preprocessing suppresses clutter components near 0 Doppler. As the preprocessing, the range-Doppler range with strong clutter level is extracted, and an M (M≧1) order approximate curve is calculated from the amplitude signal of the fast-time axis for each cell of the slow-time axis, and the original 2. The radar apparatus according to claim 1, wherein the amplitude of the clutter amplitude signal is flattened by dividing the signal. The first discrimination processing unit calculates the maximum amplitude value of the Doppler axis for each cell of each range axis in the range-Doppler signal before applying the preprocessing, and the maximum amplitude value exceeds a predetermined threshold for clutter intensity. 2. The radar device according to claim 1, wherein detection of the range cell is prohibited. The first discrimination processing unit adds the number of Doppler axis detections for each range axis cell for the CFAR detection results of the third region, and prohibits detection for range cells for which the added number of detections exceeds a predetermined threshold. The radar device according to claim 1. 2. The first discrimination processing unit deletes detection cells in which the number of detection cells within a predetermined range-Doppler range centered on each detection cell is below a predetermined threshold in the CFAR detection results of the third region. radar equipment. 7. The radar device according to claim 6, wherein the first discrimination processing unit further integrates cells within a predetermined range-Doppler range into one cell, centering on the remaining detection cell. The first discrimination processing unit adds the number of Doppler axis detections for each cell of each range axis for the CFAR detection results of the third region, and for range cells where the added number of detections exceeds a predetermined threshold, the number of detections is set as a target. 2. The radar device according to claim 1, wherein the radar device outputs a range cell. The radar device according to claim 1, wherein the second discrimination processing unit outputs a smoothed value other than velocity 0 as a target, using a smoothed value or a predicted value subjected to correlation processing and tracking processing using the observed value. Extract the CPI (Coherent Pulse Interval) signal from the received signal of the transmitted pulse,
Output the range-Doppler range from the CPI signal,
Performing pre-processing to suppress clutter components within the range-Doppler range, performing CFAR (Constant False Alarm Rate) processing to extract range-Doppler cell reflection points where the target exists;
Performing a first discrimination process to discriminate between a target and clutter from the output of the CFAR process,
Output the observed value by CPI from the discrimination result of the target and clutter,
Correlation processing and tracking processing are performed on the observed values at a predetermined period for the entire space within the search range,
performing a second discrimination process to suppress false detections from the outputs of the correlation process and the tracking process, discriminate between clutter and the target, and output a target signal;
In the CFAR processing, after performing the preprocessing, the area is divided into a first area near 0 Doppler and a second area other than that, and one-dimensional or two-dimensional first type CFAR is applied to the first area. , in the second region, a one-dimensional or two-dimensional second type CFAR is applied,
The first discrimination process performs a discrimination process for clutter and a target on the detected value of a third area that is an integration of the first area and the second area,
The second discrimination process is a radar signal processing method that discriminates the target body or blade using the results of the correlation process and tracking process.

Images