Radar system and radar signal processing method

Abstract

To allow a radar system to be implemented with a small processing scale even at the time of long-term integration and to reduce the integration loss of phase fluctuation due to range walk and Doppler walk.SOLUTION: A radar system according to an embodiment divides a fast-time axis into Mf (Mf≥2, Mf×Lf=Nf) for each cell Lf (Lf≥1), using signals of Nf (Nf≥1) cells that has transmitted and received modulated long pulses for the fast-time axis, performs a demodulation process by reference signals using Ns different speeds on the fast-time axis in each division unit, obtains Ms different ambiguity functions of Ns×Lf cells, then performs a second FFT process and arrangement of results of a first FFT process for each of the Ns×Lf cells of the Mf different ambiguity functions for each cell (nf=1 to Nf) of the fast-time axis of the ambiguity functions, detects a target using the arranged results, and achieves long-time integration by the FFT process of a small number of points (data length) using a two-stage FFT.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] This embodiment relates to a radar system and a radar signal processing method.

Background Art

[0002] In a radar system that uses continuous waves to detect small targets at long distances, when long-time integration is performed, the number of FFT points per time increases too much, the processing scale increases, and it becomes impossible to implement. Also, in the case of long-time observation, due to phase fluctuations caused by range walk and Doppler walk, a large amount of integration loss occurs.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Non-Patent Document 2

Non-Patent Document 3

Non-Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0004] As described above, conventional radar systems have two problems: when long-term integration is performed, the number of FFT points in a single operation increases too much, increasing the processing scale and making implementation impossible; and during long-term observation, significant integration losses occur due to phase fluctuations caused by range walks and Doppler walks.

[0005] The objective of this embodiment is to provide a radar system and radar signal processing method that can be implemented with a small processing scale even during long-term integration, and that can reduce integration losses due to phase fluctuations caused by range walks and Doppler walks. [Means for solving the problem]

[0006] To solve the above problems, according to the embodiment, the system is configured as follows. (1) Using the signals of Nf (Nf≧1) cells that transmit and receive modulated long pulses on the fast-time axis, the fast-time axis is divided into Mf (Mf≧2, Mf×Lf=Nf) cells for each Lf (Lf≧1) cell. Demodulation is performed on each division unit using a reference signal with speeds of Ns on the fast-time axis to obtain Mf ambiguity functions for Ns×Lf cells. Next, the results of the first FFT processing are performed on each of the Mf ambiguity function Ns×Lf cells, and the results are arranged by performing a second FFT on each cell (nf=1~Nf) on the fast-time axis of the ambiguity function. The result of this arrangement is then used for target detection.

[0007] In other words, the radar system with configuration (1) achieves long-time integration by performing FFT processing with a small number of points (data length) using a two-stage FFT.

[0008] (2) In the radar system with the configuration of (1), the array result obtained by the second FFT processing is used to tentatively detect a target, the signal output of ±Q cells is extracted, mainly from Doppler cells that exceed a predetermined amplitude threshold, shifted to Doppler 0, and the rest are filled with zeros to generate a signal, which is then subjected to inverse FFT processing, and the complex conjugate value of the processing result is used as a correction coefficient to correct the signal on the fast-time axis (PGA (Phase gradient autofocus) processing).

[0009] In other words, in the radar system with configuration (2), in addition to the functions of configuration (1), a preliminary detection is performed using CFAR, etc., and the integration loss is reduced by performing PGA processing on the preliminary detection signal to detect the target. [Brief explanation of the drawing]

[0010] [Figure 1] Figure 1 is a block diagram showing the configuration of the transmission and reception systems of a radar system according to the first embodiment. [Figure 2] Figure 2 is a flowchart showing the processing flow in the radar system according to the first embodiment. [Figure 3] Figure 3 shows the process of calculating the ambiguity function by correlation processing and the continuous wave transmission waveform in normal processing, in comparison with the radar system according to the first embodiment. [Figure 4] Figure 4 is a diagram illustrating the two-stage FFT processing in a segmented unit applied to the radar system according to the first embodiment. [Figure 5] Figure 5 shows the arrangement of the second FFT processing results for each bank of the output of the first FFT processing in the two-stage FFT processing shown in Figure 4. [Figure 6] Figure 6 shows the calculation of the ambiguity function by correlation processing with the continuous wave transmission waveform in a two-stage FFT processing at the division unit shown in Figure 4. [Figure 7] Figure 7 shows how the two-stage FFT processing between the divisions shown in Figure 6 is integrated to output the target information. [Figure 8]FIG. 8 is a block diagram showing the configuration of a transmission system and a reception system of a radar system according to the second embodiment. [Figure 9] FIG. 9 is a flowchart showing the processing flow in the radar system according to the second embodiment. [Figure 10] FIG. 10 is a flowchart showing the flow of PGA processing for correcting the integrated loss of phase fluctuations in the radar system according to the second embodiment. [Figure 11] FIG. 11 is a waveform diagram showing the state of reducing the integrated loss of phase fluctuations by the PGA processing shown in FIG. 10. [Figure 12] FIG. 1第十二図は、図6に示す分割間の2段FFT処理を統合し、図11に示すPGA処理により得られる検出目標を出力する様子を示す図である。

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments will be described with reference to the drawings.

[0012] (First Embodiment) Hereinafter, a radar system according to the first embodiment will be described with reference to FIGS. 1 to 6.

[0013] FIG. 1 is a block diagram showing the configuration of a radar system according to the first embodiment. In FIG. 1, in the transmission system, a transmission type signal is generated by a signal generator 11, a modulation signal is generated from the transmission type signal by a modulator 12, the modulation signal is converted into a high-frequency signal by a frequency converter 13, and then pulse-modulated by a pulse modulator 14 to generate a radar signal by a long pulse, which is sent into space from a transmission antenna 15.

[0014] On the one hand, the receiving system receives the radar reflection signal with the receiving antenna 16, frequency-converts the received signal to the baseband with the frequency converter 17, and converts it to a digital signal with the AD converter 18. Next, in the first FFT processor 19, the fast-time axis is divided, FFT processing is performed in the divided units, then it is divided at a predetermined range frequency by the range frequency divider 20, and a reference signal for each range frequency is generated by the reference signal generator 21. Subsequently, the ambiguity function calculator 22 calculates the ambiguity function for each divided range frequency, the second FFT processor 23 performs FFT processing on the fast-time axis, the target detector 23 detects the target by CFAR (see Non-Patent Document 2) or the like, and outputs the information of the detected target.

[0015] FIG. 2 is a flowchart showing the flow of the continuous wavelength time integration process in the radar system according to the first embodiment. First, a continuous wave radar signal with a long pulse is transmitted, and when its reflection signal is received and input (step S11), the fast-time axis is divided, FFT processing is performed in the divided units (step S12), it is divided at a predetermined range frequency (step S13), and a reference signal for each range frequency is generated (step S14). Here, range compression based on the reference signal is performed (step S15), and the compression result is saved (step S16). This process is sequentially changed in speed (steps S17, S18) and executed in the divided units of the range frequency (steps S19, S20).

[0016] Subsequently, the ambiguity function is calculated for each divided range frequency and FFT processing of the fast-time axis is performed (step S21), the FFT processing results are sequentially replaced (step S22), the Doppler cell is sequentially changed (steps S23, S24), the target is detected by CFAR, and the information of the detected target is output (step S25).

[0017] The above continuous wavelength time integration process will be described with reference to FIGS. 3 to 7.

[0018] Figure 3 shows the calculation of the ambiguity function by correlation processing with the continuous wave transmission waveform in normal processing, in comparison with the radar system according to the first embodiment; Figure 4 is a diagram for explaining the two-stage FFT processing in division units applied to the radar system according to the first embodiment; Figure 5 shows the arrangement of the second FFT processing results for each bank of the output of the first FFT processing in the two-stage FFT processing shown in Figure 4; Figure 6 shows the calculation of the ambiguity function by correlation processing with the continuous wave transmission waveform in the two-stage FFT processing in division units shown in Figure 4; and Figure 7 shows the output of target information by integrating the two-stage FFT processing between divisions shown in Figure 6.

[0019] First, Figure 3(a) shows the transmitted pulse waveform using a long pulse, Figure 3(b) shows the received pulse waveform using a long pulse, Figure 3(c) shows the reference signal corresponding to the search Doppler 1 to Ns, and Figure 3(d) shows how the target is detected using the ambiguity function and distance and velocity are calculated.

[0020] As shown in Figure 3, a reference signal is generated using the transmitted modulated signal and a Doppler probe for searching the target velocity range. This signal is then correlated with the reference signal on the range axis, where the velocity is sequentially changed along the fast-time axis and the Doppler probe axis. An ambiguity function (see Non-Patent Literature 4) is calculated from this reference signal, and once the target is detected using CFAR (see Non-Patent Literature 2), the distance and velocity can be output.

[0021] However, when performing correlation processing, it is necessary to convert to the fast-time frequency axis by FFT processing of the long pulse signal and the reference signal. However, in the case of long pulses, the number of FFT points becomes enormous, increasing the processing scale and potentially making implementation impossible. To address this, the two-stage FFT method shown in Figures 4 and 5 is applied. This involves dividing the fast-time axis into Mf cells for each Lf cell as shown in Figure 4(a), performing FFT processing with a small number of points (Lf) for each division unit as shown in Figure 4(b), rearranging the division units as shown in Figure 4(c), and then performing a second FFT processing on the rearranged result as shown in Figure 4(d) to obtain the FFT processing result for the total Nf points. This is equivalent to performing the FFT butterfly operation in two stages. For example, cell #l (l=1~Lf) shown in Figure 5(a) is divided into Mf cells (total number of cells Nf = Lf × Mf) by performing a two-stage FFT processing (arranging the second FFT processing result for each bank of the output of the first FFT processing). This method allows the FFT calculation of Nf points to be performed using an FFT calculation of up to Lf points (Lf = Nf / Mf).

[0022] Now, referring to Figure 4, we formulate the two-stage FFT. First, the input signal sig(tf,tm) is expressed as follows.

[0023]

number

[0024]

number

[0025] Next, as shown in Figure 5, a second FFT process is performed on the fast-time axis for each division unit.

[0026]

number

[0027] When the results of this FFT are arranged for each bank of the first FFT as shown in Figure 4, the same result as the FFT of the entire fast-time axis is obtained. This is shown in Figure 5.

[0028] Next, we quantify the computational scale of the FFT for all points and the two-stage FFT.

[0029]

number

[0030]

number

[0031] From the above, it can be seen that the computational complexity is the same. However, the two-stage FFT method can reduce the number of FFT points by the number of divisions, so it can be said to be an advantageous method when there are constraints on the number of FFT points.

[0032] Here, since the reference signal is calculated from the transmitted modulated signal and the velocity (Ns point) within the Doppler velocity range where the target exists, Ns calculations are required when processing the reference signal with a two-stage FFT.

[0033] First, a reference signal is generated for correlation processing using the transmitted modulated signal. The Doppler pulse output from the Doppler pulse train is used as the reference signal.

[0034]

number

[0035] The set reference signal pulse train length is Nfp, and the reference signal is zero-padded to make the received distance cell length Nf.

[0036]

number

[0037]

number

[0038] On the other hand, the received signal during the ranging period can be expressed by the following equation.

[0039]

number

[0040]

number

[0041]

number

[0042]

number

[0043]

number

[0044] Figures 6 and 7 show the overall processing. Figure 6(a) is the transmitted pulse (long pulse), Figure 6(b) is the received pulse (long pulse), Figure 6(c) is the received pulse (long pulse) after two-stage FFT processing, and Figure 6(d) is the Doppler 1 to Mf reference signal obtained by two-stage FFT processing of the transmitted pulse (long pulse). Figure 7(a) shows the correlation processing by calculating the ambiguity function, where a second FFT processing (Mf point) is performed between each division unit, and Figure 7(b) shows the target detection by synthesizing the processing results of the second FFT processing between divisions.

[0045] In other words, as is clear from Figures 6 and 7, the target distance can be calculated by detecting the threshold of the ambiguity function srng using CFAR or the like, and converting the time axis to the distance axis. The velocity can be extracted from the Doppler cell of the detected reference signal.

[0046] (Second embodiment) A radar system according to the second embodiment will be described below with reference to Figures 8 to 12.

[0047] In the first embodiment, a method for creating a range (fast-time)-Doppler axis ambiguity function using a two-stage FFT was described for processing long pulses. In the case of long pulse processing, phase fluctuations may occur within the fast-time axis, resulting in processing loss. In this embodiment, an example of applying a method similar to PGA (Phase gradient autofocus, see Non-Patent Literature 3), an autofocus method for synthetic aperture processing, as a correction method is described. Figure 8 is a block diagram showing the configuration of the transmission and reception systems of the radar system according to the second embodiment, Figure 9 is a flowchart showing the processing flow in the radar system according to the second embodiment, Figure 10 is a flowchart showing the flow of PGA processing to correct the integral loss of phase fluctuations in the radar system according to the second embodiment, Figure 11 is a waveform diagram showing how the integral loss of phase fluctuations is reduced by the PGA processing shown in Figure 10, and Figure 12 is a diagram showing how the two-stage FFT processing between divisions shown in Figure 6 is integrated and the detection target obtained by the PGA processing shown in Figure 11 is output.

[0048] In Figures 8 and 9, the difference from the configuration of the first embodiment shown in Figures 1 and 2 is that after the second FFT processor 23 performs FFT processing on the fast-time axis, the target is tentatively detected by the target tentative detector 25 using CFAR or the like (step S26), the PGA processor 26 performs PGA processing on the fast-time axis for each Doppler cell of the tentative target to correct the integral loss of phase fluctuations (steps S27-S29), the target is detected by the target detector 24, and the detected target information is output (step S30).

[0049] Here, the phase correction process of the PGA processor 26 described above, as shown in Figure 10, first inputs the fast-time axis signal (step S41), extracts the peak value (step S42), rearranges the range axis (zero-shifts the peak value) (step S43), and performs window function multiplication (±R cell) and zero-padding (step S44). Subsequently, a second FFT process is performed (step S45), a phase correction value is calculated from the correction coefficient (step S46), the phase gradient is corrected with the phase correction value, and then an inverse FFT process is performed (step S47) to output the PGA processed result with the main lobe improved.

[0050] Figure 11 is a diagram that specifically explains the phase correction shown in Figure 10. As shown in Figure 11, we will explain one Doppler cell among the tentatively detected Pt combinations in the range-Doppler axis results (range-Doppler data) of the ambiguity function. The same process can be performed for the other tentatively detected Doppler cells.

[0051] From the signal on the provisionally detected range axis (Nf = Mf × Lf cell), the maximum value (peak value) that exceeds a predetermined amplitude threshold is extracted (Figure 11(a1), (b1)). Next, in order to remove the phase gradient of the maximum value relative to the Doppler axis and extract only the phase fluctuation, the signal on the range axis is rearranged so that the maximum value is shifted to the center of the range axis (corresponding to 0 of the FFT frequency) (Figure 11(a2), (b2)). Next, in order to remove the oscillation component of the phase shift and obtain a stable corrected component, a window function is multiplied by ±R (R≧1) cells centered on the maximum value (0 range), and a signal s0 is generated by zero-filling the area outside the window function, and this signal is subjected to FFT processing (Figure 11(c)).

[0052]

number

[0053] The correction amount Wc(tf), which is the inverse characteristic of this signal s0(tf), is used as the correction value for the fast-time axis signal to correct the input signal, and the range axis signal is obtained by inverse FFT processing (Figure 11(d)).

[0054]

number

[0055] Figure 12(b) shows the result of PGA processing on the provisionally detected ambiguity function calculation result shown in Figure 12(a). That is, as is clear from Figure 12, the target distance of the PGA processed output can be calculated by detecting the threshold of the ambiguity function srng using CFAR or the like, and converting the time axis to the distance axis. The velocity can be extracted from the Doppler cell of the detected reference signal.

[0056] It should be noted that the present invention is not limited to the above embodiments, and the components can be modified and implemented in practice without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiments. Moreover, components from different embodiments may be appropriately combined. [Explanation of Symbols]

[0057] 11...Signal generator, 12...Modulator, 13...Frequency converter, 14...Pulse modulator, 15...Transmitting antenna, 16...Receiving antenna, 17...Frequency converter, 18...AD converter, 19...First FFT processor, 20...Range frequency divider, 21...Reference signal generator, 22...Ambiguity function calculator, 23...Second FFT processor, 24...Target detector, 25...Target provisional detector, 26...PGA processor.

Claims

1. A transmitting and receiving means for transmitting and receiving modulated long pulses on the fast-time axis, A division means that uses the signals of the transmitted and received Nf (Nf≧1) cells to divide the fast-time axis into Mf (Mf≧2, Mf×Lf=Nf) cells for each Lf (Lf≧1), A reference signal generation means generates a reference signal using Ns different speeds on the fast-time axis for each division unit, Demodulation means that demodulates the signals of Mf cells, which are divided into Mf cells for each Lf (Lf≧1) cell along the fast-time axis, using the reference signal to obtain the ambiguity function of Ns×Lf cells according to Mf, A first FFT processing means performs a first FFT (Fast Fourier Transform) process on each of the Ns × Lf cells of the aforementioned Mf ambiguity function, A second FFT processing means that processes the results of the first FFT processing for each cell (nf = 1 to Nf) on the fast-time axis of the ambiguity function and arranges them, A detection means for detecting a target using the cells arranged by the second FFT process, A radar system equipped with the following features.

2. The radar system according to claim 1, wherein the detection means provisionally detects a target using the array result obtained by the second FFT processing, extracts the signal output of ±Q cells centered on the Doppler cell that exceeds a predetermined amplitude threshold, shifts it to Doppler 0, generates a signal with zeros filled in the rest, performs FFT processing on the generated signal, and corrects the fast-time axis signal using the complex conjugate value of the processing result as a correction coefficient in a PGA (Phase gradient autofocus) processing.

3. On the fast-time axis, modulated long pulses are transmitted and received. Using the signals from the transmitted and received Nf cells (Nf≧1), the fast-time axis is divided into Mf cells (Mf≧2, Mf×Lf=Nf) for each Lf cell (Lf≧1), In each division unit, a reference signal is generated using Ns different speeds on the fast-time axis. The fast-time axis is divided into Mf cells (Lf≧1) and the signals of the Mf cells are demodulated using the reference signal to obtain the ambiguity function of the Mf-type Ns×Lf cells. For each of the Ns × Lf cells of the aforementioned Mf ambiguity function, a first-stage Fast Fourier Transform (FFT) is performed. The results of the first FFT process are then processed using a second FFT for each cell on the fast-time axis of the ambiguity function (nf = 1 to Nf) and arranged. The target is detected using the cells arranged by the second FFT process. Radar signal processing method.

4. The radar signal processing method according to claim 3, wherein a target is provisionally detected using the array result obtained by the second FFT processing, the signal outputs of ±Q cells are extracted, mainly from Doppler cells that exceed a predetermined amplitude threshold, shifted to Doppler 0, and the rest are padded with zeros to generate a signal, which is then processed with FFT, and a Phase Gradient Autofocus (PGA) process is performed to correct the fast-time axis signal using the complex conjugate value of the processing result as a correction coefficient, and target detection is performed on the PGA processing result.

Images