JP2010243247A - Signal processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that a plurality of adjacent targets cannot be detected because input signal characteristics do not match with CFAR (Constant False Alarm Rate) circuit characteristics or incorrect detection is caused due to the failure in suppressing clutter in a conventional CFAR circuit. <P>SOLUTION: A rise in noise level due to a specific amplitude is suppressed by averaging with the maximum value within each reference cell in the CFAR circuit excluded. When there is a broad signal such as clutter, the noise level rises and the incorrect alarm is suppressed by using GO-CFAR. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a signal processing apparatus having a CFAR (Constant False Alarm Rate) circuit mounted on a radar apparatus and used for target detection.

  The CFAR circuit is used in a radar signal processing device that suppresses unnecessary signals such as clutter and noise from a received signal of a radar receiver and detects a required target signal. The CFAR circuit determines a threshold value (threshold) for signal detection with respect to an input signal, and performs target detection while keeping the false alarm probability constant. The CFAR circuit is mounted on a radar signal processing device such as a search or tracking radar, and various proposals have been made regarding its configuration.

  As an example of a CFAR circuit, a technique for switching between GO (Greatest Of) -CFAR and CA (Cell Averaging) -CFAR according to the characteristics of an input signal, and a technique for switching to another CFAR are known (for example, Patent Documents). 1, see Patent Document 2).

Japanese Patent Laid-Open No. 3-138585 (page 4, FIG. 1)

Japanese Patent Laid-Open No. 9-145829 (page 11, FIG. 1)

  In general, in aircraft-mounted and vehicle-mounted radar devices that require miniaturization, it is difficult to provide a characteristic determination circuit for determining unknown input signal characteristics and multiple CFAR circuits due to dimensional constraints. The fact is that only one CFAR circuit closest to the signal characteristics is mounted.

  However, in a conventional small radar device, the input signal characteristic and the CFAR circuit characteristic do not match, and therefore, there are cases where a plurality of adjacent targets cannot be detected, or clutter and interference wave noise cannot be suppressed and the target is erroneously detected. there were.

  The present invention has been made to solve such a problem, and suppresses unnecessary waves such as clutter and interference wave noise, and lowers the frequency of erroneous detection of targets and the undetected occurrence of multiple adjacent targets. Objective.

  The signal processing apparatus according to the present invention includes at least two reference cells that respectively extract amplitude values of signals on both sides of the selected target cell in the range bin with respect to the input signal, and at least integrate the signals extracted from the respective reference cells. Two integrators, at least two maximum value detectors for detecting the maximum values from the signals extracted from the respective reference cells, and the maximum values detected by the respective maximum value detectors from the respective outputs of the respective integrators. At least two arithmetic units excluding the values, at least two average value calculators for calculating an average value according to the number of cells of each reference cell from outputs of the respective arithmetic units, and on both sides of each target cell. A comparator for comparing the outputs of the two average value calculators, and a multiplier for multiplying the output of the comparator by a constant. A CFAR (Constant False Alarm Rate) circuit comprising a target detector that compares the amplitude value of the eye cell with the target cell and determines that the target cell is the target when the amplitude value of the target cell is larger. It is a thing.

  According to the present invention, the noise level rise due to the specific amplitude can be suppressed by averaging except the maximum value in each reference cell in the CFAR circuit. In addition, when a broad signal such as clutter exists, the use of GO-CFAR can increase the noise level and suppress false alarms.

It is a figure for demonstrating the structure of the GO-CFAR circuit by Embodiment 1 of this invention. It is a figure for demonstrating operation | movement of the GO-CFAR circuit by Embodiment 1 of this invention. It is a figure for demonstrating operation | movement of the GO-CFAR circuit of a comparative example. It is a figure for demonstrating operation | movement of the SO-CFAR circuit of a comparative example.

Embodiment 1 FIG.
The CFAR circuit according to the first embodiment of the present invention will be described below with reference to the drawings. The CFAR circuit according to the first embodiment is provided in a target detection signal processing apparatus and includes a maximum value detection circuit in a reference cell in a range bin. FIG. 1 is a diagram illustrating a configuration of a GO-CFAR type CFAR circuit according to the first embodiment.

  The received signal transmitted from the radar transmitter, reflected by the target and input to the radar receiver is converted into a digital signal and then input to the radar signal processing device. In the radar signal processing apparatus, clutter, which is a reflected wave from the ground, mountains, sea, rain, clouds, etc., is removed by a clutter suppression circuit. The received signal from which the clutter has been removed is detected in amplitude by a linear detector, then subjected to processing for improving S / N in an integration circuit, and then input to a target detection signal processor for target detection. (For example, refer to Patent Documents 1 and 2). In this target detection signal processing apparatus, an input digitized reception signal is stored in a range bin configured in a memory and then input to the CFAR circuit according to the first embodiment, so that a necessary threshold value is obtained. Based target detection is performed.

  In FIG. 1, the CFAR circuit according to Embodiment 1 includes two systems of maximum value detectors 4a and 4b, two systems of integrators 3a and 3b, two systems of arithmetic units 5a and 5b, and an average value of two systems. Calculators 6a and 6b, a comparator 7 and a detector 8 are provided. The received signal input to the CFAR circuit sets a moving window for the reception time divided by the range bin interval, and sets the attention cell and the attention cell for the amplitude value of the signal stored in each range bin within the moving window range. Reference cells 2a and 2b adjacent to each other are set. Each of the reference cell 2a and the reference cell 2b is composed of a plurality of cells whose number of cells is Nc. The reference cell 2a and the reference cell 2b take out the amplitude level of the signal from the input signal as an amplitude value in order to calculate the noise level. The signals extracted by the reference cell 2a and the reference cell 2b are respectively integrated by the integrator 3a and the integrator 3b. Further, the maximum value of each reference cell is detected by the maximum value detector 4a and the maximum value detector 4b from the signals extracted by the reference cell 2a and the reference cell 2b.

  The arithmetic units 5a and 5b subtract the maximum values of the reference cells detected by the maximum value detector 4a and the maximum value detector 4b, respectively, from the integration results of the integrators 3a and 3b. . The average value calculator 6a and the average value calculator 6b calculate the average value excluding the maximum value from the reference cell by dividing the calculated values by the calculators 5a and 5b by the number of cells Nc. The comparator 7 compares the arithmetic values of the arithmetic units 6a and 6b, whose average values are calculated by the average value calculator 6a and the average value calculator 6b, respectively, and outputs the larger arithmetic value based on the comparison result. . The detector 8 calculates a value (B * K) obtained by multiplying the output B of the comparator 7 by a multiplier (coefficient) K input from an external circuit of the detector 8 as a threshold level. The detector 8 compares the threshold value (B * K) obtained by this multiplication with the output A of the target cell 1 to be sequentially selected, and when the output A of the target cell 1 is larger than the threshold value (B * K). The output A is output as a target amplitude value.

  Next, the operation of each unit will be described with respect to an input signal in which two targets and clutter that are close on the frequency axis exist. FIG. 2 is a diagram for explaining the operation of each part of the GO-CFAR type CFAR circuit according to the first embodiment.

  In the figure, for the input signal 9, the arithmetic unit 6 a in FIG. 1 outputs an output signal 10. The calculator 6b of FIG. 1 compares the input signal 10 and the output signal 11 and outputs the larger one as the output signal 12. The detector 8 shown in FIG. 1 compares the input signal 12 multiplied by the multiplier K with the signal level of the target cell 1 shown in FIG. 1, and outputs the target signal 13 when the signal level of the target cell is high. Here, for example, when the multiplier K is set to 3, the clutter shown in FIG. 2 is smaller than three times the output signal 12 in the cell of interest, and is not detected as a target. On the other hand, the target signal 13 shown in FIG. 2 is larger than three times the output signal 12 in the target cell, so that it is reliably detected as a target.

  That is, in the example of FIG. 2, it is shown that clutter having an amplitude level larger than the target level and spreading in the frequency axis direction is not erroneously detected as a target, but two adjacent targets are detected as the target signal 13.

  Next, regarding the CFAR circuit in the case where the maximum value detectors 4a and 4b and the arithmetic units 5a and 5b in FIG. 1 are not provided and the outputs of the integrators 3a and 3b are directly input to the arithmetic units 6a and 6b, A corresponding comparative example will be described. FIG. 3 is a diagram for explaining the operation of each part of the GO-CFAR circuit of the comparative example.

  In the figure, the comparator 7 of the comparative example outputs a signal 15 with respect to the input signal 14. The detector 8 compares the signal 15 multiplied by the coefficient K with the signal level of the cell of interest 1 shown in FIG.

  The signal 15 in FIG. 3 has a higher amplitude level than the signal 13 in FIG. 2 because the maximum value signal is not subtracted. Thus, although the clutter is not detected as a target, the proximity target that can be detected as the target signal 13 in FIG. 2 cannot be detected in FIG.

  Next, the operation in the case of a conventional SO (Smallest Of) -CFAR in which the comparator 7 of FIG. 1 does not output a high input level but outputs a low one will be described. FIG. 4 is a diagram for explaining the operation of the SO-CFAR circuit of the comparative example.

  In response to the input signal 16, the SO-CFAR comparator 7 outputs a signal 17. The detector 8 compares the signal 17 multiplied by the coefficient K with the signal level of the cell of interest 1 in FIG.

  The signal 17 of FIG. 4 has a lower amplitude level than the signal 12 of FIG. FIG. 4 shows that the proximity target can be detected as the target signal 18, but the clutter is erroneously detected as the target signal 19.

  Needless to say, the GO-CFAR circuit according to the first embodiment exhibits the same effect not only on the frequency axis but also on the time axis.

  As described above, the target detection signal processing apparatus according to the first embodiment includes at least two reference cells that respectively extract the amplitude values of the signals on both sides of the selected target cell in the range bin with respect to the input signal, and each of the reference cells. At least two integrators for integrating the signals taken out from the reference cells, at least two maximum value detectors for detecting the maximum values from the signals taken out from the respective reference cells, and the respective outputs from the outputs of the respective integrators. At least two arithmetic units each excluding the maximum value detected by the maximum value detector, and at least two average value calculators for calculating an average value according to the number of cells of each reference cell from respective outputs of the respective arithmetic units A comparator for comparing the outputs of the two average value calculators on both sides of each cell of interest, and A target detector that multiplies a force by a constant, compares the multiplied value with the amplitude value of the cell of interest, and determines that the cell of interest is a target when the amplitude value of the cell of interest is greater. According to this embodiment including a constant false alarm rate (CFAR) circuit, the noise level rise of a specific amplitude is suppressed by averaging except for the maximum value in each reference cell in the GO-CFAR circuit. In addition, when there is a broad signal such as clutter, the noise level rises and false alarms can be suppressed. Furthermore, performance improvement is expected by adding a simple circuit to the conventional CFAR circuit.

  1 cell of interest, 2a reference cell, 2b reference cell, 3a integrator, 3b integrator, 4a maximum value detector, 4b maximum value detector, 5a calculator, 5b calculator, 6a average value calculator, 6b average value calculation , 7 comparator, 8 detector, 9 input signal, 10 noise level, 11 noise level, 12 noise level, 13 target signal, 14 input signal, 15 noise level, 16 input signal, 17 noise level, 18 target signal, 19 False detection target signal.

Claims (1)

At least two reference cells that respectively extract the amplitude values of the signals on both sides of the selected cell of interest in the range bin for the input signal;
At least two integrators for integrating the signals extracted from the reference cells, respectively;
At least two maximum value detectors for detecting a maximum value from the signal extracted from each of the reference cells;
At least two computing units each excluding the maximum value detected by each maximum value detector from the respective output of each integrator;
At least two average value calculators for calculating the average value according to the number of cells of each reference cell from the respective outputs of the respective arithmetic units;
A comparator that compares the outputs of the two average value calculators on both sides of each cell of interest;
A target detector that multiplies the output of the comparator by a constant, compares the multiplied value with the amplitude value of the cell of interest, and determines that the cell of interest is the target if the amplitude value of the cell of interest is greater When,
A signal processing apparatus provided with a CFAR (Constant False Alarm Rate) circuit.

Images