Radar Systems
The radar system employs threshold-based peak selection to reduce processing load and suppress false images, enhancing target detection accuracy by simplifying calculations and eliminating sidelobe interference.
Patent Information
- Application Number
- JP2023202069
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-11-29
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2039-07-08
AI Technical Summary
Conventional sidelobe cancellers in radar systems face high processing loads and sensitivity issues when attempting to suppress sidelobes, and struggle to eliminate false images caused by factors other than interference waves.
A radar system that uses a transmitter, receiving antenna array, and signal processing units to generate distance and azimuth spectra, extract intensity peaks, and apply threshold-based selection to identify main peaks for target detection, reducing the need for heavy calculation and effectively suppressing false images.
This approach reduces processing load and effectively suppresses false images, allowing for accurate target detection with minimal computational effort while addressing sidelobe interference.
Smart Images

Figure 0007588704000001 
Figure 0007588704000002 
Figure 0007588704000003
Abstract
Description
[Technical field]
[0001] This invention relates to a radar system for measuring distance and direction to a target. [Background technology]
[0002] Many radar systems have been proposed that use antenna arrays to measure the distance and direction to a target. In these radar systems, the incoming wave from the target is the desired signal, but various factors can cause interference waves to be superimposed, resulting in the detection of false images. In radar systems using antenna arrays, a method is adopted in which a phase shifter is used to align the phase of the output waveforms of each element antenna to the desired receiving beam direction, thereby maximizing the composite output (response characteristic) of the antenna array in that direction (the so-called beamforming method).
[0003] Mathematically, the combined output E of the antenna array is obtained by multiplying the output waveform function of each element antenna by a weighting coefficient w (complex number) that gives the amount of phase shift and adding them up. By giving the weighting coefficient w as a function w(θ) of the azimuth θ, the combined output E of the antenna array can also be expressed as a function E(θ). The combined output E(θ) is a complex function, and the absolute ||E(θ)|| is mapped for various angle values θ, which is called the directional pattern of the antenna array, and the area around its maximum value is called the main lobe (main beam). On the other hand, it is said that the weighting coefficient w should be selected so that the response characteristics at other angle positions than the main lobe are as small as possible, ideally close to zero, but mathematical principles mean that local maximum values of the response characteristics occur at the other angle positions. This is called a side lobe.
[0004] If jamming waves arrive from the direction of the side lobe, they will cause radar false images, and so various side lobe cancellers have been proposed to suppress these side lobes. In a side lobe canceller, a weighting coefficient of a function E that gives the combined output of the antenna array is determined by calculation so that the antenna gain in the direction from which the jamming waves arrive is reduced. As a method for determining this weighting coefficient w, a method that converges to an optimal value by successively updating the value of the weighting coefficient w for each sample of the received signal is often adopted (for example, Patent Document 1 and Non-Patent Document 1). [Prior art documents] [Patent documents]
[0005] [Patent Document 1] JP 2006-162397 A [Non-patent literature]
[0006] [Non-Patent Document 1] "Adaptive Sidelobe Suppression Method for High Gain Arrays", IEICE Transactions on Information and Communication Engineers, Vol. J86-B, No.5, pp.790-797, May 2003 Summary of the Invention [Problem to be solved by the invention]
[0007] However, the above-mentioned conventional side lobe canceller has the following problems. (1) The load of repeated calculation of the weighting coefficient w for converging the output value of the array antenna to the optimum value increases, and the processing time becomes longer. (2) If an attempt is made to significantly reduce the side lobe level, the main lobe level will also be reduced, which may result in a decrease in sensitivity. (3) It is difficult to eliminate false images caused by factors other than interference waves.
[0008] An object of the present invention is to provide a radar system that requires only a small computational processing load and can effectively suppress false images caused by factors other than jamming waves. [Means for solving the problem]
[0009] In order to solve the above problems, the radar system of the present invention includes a transmitter that transmits search radio waves for searching for a target as transmission waves, a receiver that has a receiving antenna array consisting of multiple element antennas and receives reflected waves of the search radio waves as received waves via the receiving antenna array, a scanner that performs beam scanning of the received waves by controlling the receiving beam direction of the receiving antenna array while sequentially changing it within a predetermined azimuth angle range, a distance spectrum generator that generates distance spectra showing the relationship between distance and reception intensity in the receiving beam direction for each direction based on the transmitted waveform signal from the transmitter and the received waveform signals of various receiving beam directions obtained by beam scanning, and a distance spectrum generator that generates a distance spectrum ... that shows the relationship between distance and reception intensity in the receiving beam direction for each direction based on the transmitted waveform signal from the transmitter and the received waveform signals of various receiving beam directions obtained by beam scanning. the signal processing unit is configured to extract received wave intensity points at the same distance from the received signal, thereby generating, for each distance, an orientation spectrum indicating the relationship between the receiving beam orientation and the receiving intensity; a peak extraction unit is configured to extract an intensity peak from each of the orientation spectra generated for each distance; a peak selection processing unit is configured to set, for each set of intensity peaks extracted for each individual orientation spectrum, a selection condition including, as a necessary condition, that the intensity value exceeds a predetermined threshold value, and to select, from the set of intensity peaks, an intensity peak that satisfies the selection condition as a main peak for identifying a target; and a target detection output unit is configured to output information on the distance and orientation indicated by the selected main peak as target detection information. Effect of the Invention
[0010] In the above configuration of the present invention, when multiple intensity peaks occur on the azimuth spectrum by distance, a selection is performed on a set of the intensity peaks, with the necessary condition being whether the intensity value exceeds a threshold, and the distance and azimuth information indicated by the main peak that satisfies the selection condition is output as target detection information. This makes it possible to replace the heavy calculation processing that was previously required when it was desired to sufficiently reduce side lobes by weighting calculation with a simple intensity peak selection process, thereby realizing a radar system with a low calculation processing load. In addition, the selection process can effectively suppress intensity peaks caused by false images generated by factors other than jamming waves. [Brief description of the drawings]
[0011] [Figure 1] FIG. 1 is a functional block diagram of a radar system according to an embodiment of the present invention. [Diagram 2] FIG. 4 is a circuit diagram showing an example of a specific configuration of a transmission unit. [Diagram 3] FIG. 2 is a block diagram showing a more specific electrical configuration of the signal processing device of the radar system shown in FIG. 1. [Figure 4] FIG. 4 is a schematic diagram showing an example of a frequency modulation pattern of a transmission wave. [Diagram 5] 4A and 4B are diagrams for explaining the concept of analyzing the frequency spectrum of a beat waveform signal. [Figure 6] FIG. 4 is a diagram conceptually showing the contents of a target list memory. [Figure 7] 4 is a flowchart showing the overall processing flow of a measurement control program. [Figure 8] 5 is a flowchart showing the flow of the direction analysis process in FIG. 4 . [Figure 9] 5 is a flowchart showing the flow of the peak analysis process in FIG. 4. [Figure 10] 5 is a flowchart showing the procedure of the artifact filtering process in FIG. 4. [Figure 11] 11 is a flowchart showing a main part of a modified example of the process in FIG. 10 . [Figure 12] FIG. 13 is a schematic diagram showing an example of an azimuth spectrum together with an example of setting an artifact search section. [Figure 13]FIG. 13 is a schematic diagram showing a state in which suppression processing has been performed on the azimuth spectrum of FIG. 12 in such a way that the intensity of the false image peak is made uniform at a constant level. [Figure 14] FIG. 13 is a schematic diagram showing the concept of effective false image peak intensity in the azimuth spectrum of FIG. 12 . [Figure 15] FIG. 15 is a schematic diagram showing a state in which suppression processing is performed in a manner that reduces the ratio between the azimuth reference level of the azimuth spectrum and the effective false image peak intensity in FIG. 14 . [Figure 16] FIG. 4 is a mapping diagram showing an example of a target detection result before false image filtering is performed. [Figure 17] FIG. 14 is a mapping diagram showing an example of a result when artifact filtering is performed by the method of FIG. 13. [Figure 18] 16 is a mapping diagram showing an example of a result when artifact filtering is performed by the method of FIG. 15. [Figure 19] FIG. 19 is a mapping diagram showing an example of a result when the reduction coefficient is set larger than that in FIG. 18. [Figure 20] FIG. 4 is a schematic diagram showing an example of a frequency modulation pattern of a transmission wave when speed detection is performed. [Figure 21] FIG. 13 is a diagram conceptually showing the contents of a target list memory when target detection output is performed by identifying speed. [Figure 22] A flow chart showing the process flow of the measurement control program when performing false image filtering after peak detection. [Figure 23] FIG. 23 is an explanatory diagram showing the concept of performing the false image filtering process in FIG. 22 on a target list memory. [Figure 24] FIG. 13 is a diagram for explaining the concept of correcting the reference level of the azimuth spectrum according to the background swell level. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0012] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. 1 is a block diagram showing the electrical configuration of a radar system 1 according to one embodiment of the present invention. The radar system 1 is configured as a device that measures the distance and direction to a target based on the principle of frequency modulated continuous wave (FMCW) radar, and has a signal processing unit 2 and a front-end unit 3 for transmitting and receiving radio waves that is connected thereto.
[0013] The front-end unit 3 includes a master IC 50M and a slave IC 50S cascade-connected to the master IC 50M via a wiring section SCL. In this embodiment, the master IC 50M and the slave IC 50S are both configured as the same type of radar front-end IC, and can be appropriately selected to function as a master IC or a slave IC by switching the setting input to a predetermined switching terminal (not shown). An example of a commercially available radar front-end IC of this kind is the AWR1243P (manufactured by Texas Instruments, USA).
[0014] The master IC 50M incorporates an oscillator circuit 54, a transmission amplifier 41, and a transmission antenna 51, which constitute a transmission section (they are also incorporated in the slave IC 50S, but are not used). The oscillator circuit 54 generates a transmission waveform signal whose frequency is continuously modulated based on a frequency instruction value input from a control circuit 56. The transmission waveform signal is amplified by the transmission amplifier 41, and then output from the transmission antenna 51 as a transmission wave (radio wave for target search). As shown in FIG. 2, the transmission antenna 51 is configured as a transmission antenna array consisting of a plurality of element antennas 51(A)-51(C). The transmission waveform signal from the oscillator circuit 54 is phase-adjusted by the phase shifter 42 for each of the element antennas 51(A)-51(C), and is superimposed on each other and transmitted as a directional transmission beam in a desired azimuth (θ). The phase shifter 42 receives a phase-shift control signal from the control circuit 56 in the IC, and plays a role of scanning the directional transmission beam within a predetermined azimuth angle range together with the control circuit 56.
[0015] On the other hand, the master IC 50M and the slave IC 50S are provided with a plurality of element antennas (receiving antennas) 52 constituting a receiving antenna array, which are distributed and incorporated therein, and a low-noise amplifier 43 that amplifies the waveform signal of the wave received by the element antennas 52 is provided as a receiving section. The waveform signal output from the low-noise amplifier 43 is mixed in a mixer 53 with the waveform signal of the transmission wave distributed from the master IC 50M via a wiring section SCL. The beat waveform signal output from the mixer 53 is then digitally converted in an A / D converter 55 to become beat waveform data X1(t)-Xn(t), which are serially output from waveform output ports WO1-WOn. The A / D converter 55 has its waveform frame capture start timing and output timing of each waveform bit controlled by a clock signal input from a clock circuit 59 via a control circuit 56.
[0016] The transmitted wave follows a linear modulation pattern with a constant frequency deviation width (occupied frequency band) Δf and period T. The frequency of the transmitted wave changes depending on the distance to the target during the period from when the transmitted wave is sent to when the reflected wave returns. As a result, when a transmitted wave sent at a certain time returns as a reflected wave, if this received wave and the transmitted wave at the time of reception are mixed, a standing wave equivalent to the difference frequency between the two is generated as a beat. Specifically, if the distance to the target is d, the frequency of the beat signal (= the difference frequency between the transmitted frequency f1 and the received frequency f2) is δf, and the speed of light is c, then δf=2 d Δf / (c T) Since c, T and Δf are constant, it can be seen that the frequency Δf of the standing wave beat signal is simply proportional to the distance d. Therefore, the distance d to the target can be directly calculated from measuring Δf. Figure 4 shows an example of a sawtooth linear modulation pattern, where the frequency change rate for each period T is always constant. The fundamental frequency of the transmission wave is appropriately set in the range of 75 GHz to 80 GHz, the frequency deviation width Δf of the transmission wave is appropriately set in the range of 300 MHz to 1.5 GHz (preferably, 300 MHz to 1 GHz), and the period T is appropriately set in the range of 0.5 msec to 10 msec (preferably, 1 msec to 5 msec).
[0017] In addition, in FIG. 1, only one set of the receiving element antenna 52, low-noise amplifier 43, mixer 53 and A / D conversion unit 55 is drawn for each master IC 50M and slave IC 50S, but in reality, multiple sets of these are provided in each IC 50M, 50S (for example, four sets in the AWR1243P).
[0018] The master IC 50M and each slave IC 50S are cascade-connected by a wiring section SCL that connects the synchronous output port SYNO and the synchronous input port SYNI provided in each IC. The wiring section SCL, together with the IC internal wiring, forms a waveform distribution line WVL and a control signal line SCL. A transmission waveform signal from the oscillator circuit 54 of the master IC 50M is input to each mixer 53 in the slave IC 50S via a buffer amplifier 57 through the waveform distribution line WVL.
[0019] Meanwhile, the control circuit 56 of each slave IC 50S receives a synchronization signal from the master IC 50M via the signal line SCL, and inputs the clock signal input from the clock circuit 59 based on the synchronization signal to the A / D conversion unit 55 while correcting the timing. As a result, each slave IC 50S (and the master IC 50M) outputs beat waveform data from the waveform output ports WO1 to WOn in a form in which the waveform frame capture start timing and the output timing of each waveform bit are synchronized. Therefore, the control circuit 56 functions as a synchronization control unit.
[0020] Next, the signal processing device 2 has the following functional blocks: a distance spectrum generating unit 60A, an orientation spectrum generating unit 60B, a peak extracting unit 60C, and a false image filtering unit 60D (functioning as a peak selection processing unit and a target detection output unit). Each functional block is characterized as follows. Distance spectrum generating unit 60A: Provided in a one-to-one correspondence with the master IC 50M and the slave IC 50S, it converts beat waveform data (signals) output from the waveform output ports WO1 to WOn of each IC into a frequency spectrum, which is a distance spectrum. In this embodiment, the entire scanning angle range of the receiving beam azimuth is divided and covered by the master IC 50M and the slave IC 50S into sub-scanning angle ranges with different central azimuths. Each distance spectrum generating unit 60A determines a weighting coefficient w(θ) for each azimuth θ by sequentially changing the receiving beam azimuth θ in the sub-scanning angle range it is responsible for, according to a known beamforming algorithm, and multiplies the beat waveform data received from the corresponding IC 50M, 50S by the weighting coefficient w(θ) and performs an addition operation. This allows the beat waveform data for each receiving beam azimuth θ to be synthesized (that is, the function of a scanning unit that performs beam scanning of the receiving wave by sequentially changing and controlling the receiving beam azimuth of the receiving antenna array in a predetermined azimuth angle range is realized). Then, by performing fast Fourier transform processing on the beat waveform data obtained for each azimuth θ within the sub-scanning angle range (i.e., based on the transmission waveform signal from the transmitter and the reception waveform signals of various reception beam azimuths obtained by beam scanning), a distance spectrum showing the relationship between distance and reception intensity in the reception beam azimuth θ is generated for each azimuth within the sub-scanning angle range.
[0021] Azimuth spectrum generation unit 60B: By extracting received wave intensity points at the same distance from the multiple distance spectra generated for each direction by each distance spectrum generation unit 60A, an orientation spectrum showing the relationship between the receiving beam orientation and the receiving intensity is generated for each distance d. Peak extraction unit 60C: Extracts intensity peaks from each of the azimuth spectra generated for each distance. False image filter unit 60D: This performs the function of the false image peak suppression unit. Specifically, for a set of intensity peaks extracted for each azimuth spectrum, a selection condition including a necessary condition that the intensity value exceeds a predetermined threshold is set, and an intensity peak that satisfies the selection condition in the set of intensity peaks is selected as a main peak for target identification (peak selection processing unit). Then, distance and azimuth information indicated by the selected main peak is output from the output port DO as target detection information (target detection output unit).
[0022] 3 is a block diagram showing the details of the electrical configuration of the signal processing unit 2. The signal processing unit 2 has a CPU 31, a RAM 32, a ROM 33, an input / output unit 34, and a bus 30 connecting them. The following programs are stored in the ROM 33. All of these programs are executed by the CPU 31 using the RAM 32 as an execution area, and embody individual function realization units. Measurement control program 33a: This is a basic program for realizing the functions of the distance spectrum generating section 60A, the azimuth spectrum generating section 60B, the peak extracting section 60C, and the false image filtering section 60D. Digital filter 33b: This takes the beat signal waveform data into the waveform filter memory 32a of the RAM 32 and performs digital filtering to remove noise derived from harmonics, etc. It is configured as a low-pass filter or a band-pass filter.
[0023] Reception beam scanning program 33c: A program that realizes part of the functions of the azimuth spectrum generating unit 60B. Beat waveform data output from the waveform output ports WO1 to WOn is received by each input port WI1 to WIn of the input / output unit 34, and is taken into the waveform memories w1 to wn of the RAM 32. Then, a weighting coefficient w(θ) for each azimuth θ is determined by sequentially changing the reception beam azimuth θ in the above-mentioned sub-scan angle range (for example, it may be read from a weighting coefficient table for each azimuth stored in the ROM 33), and in the waveform synthesis / scanning memory 32b, the beat waveform data received from the corresponding ICs 50M, 50S is multiplied by the weighting coefficient w(θ) and an addition operation is performed to synthesize beat waveform data for each azimuth θ. FFT module 33d: The beat waveform data for each direction shown in FIG. 5 is converted into a frequency spectrum Spm using the FFT processing memory 32h in the RAM 32, and a distance spectrum showing the relationship between the distance and the reception intensity in the receiving beam direction θ is generated for each direction. Azimuth analysis program 33e: A program that realizes part of the functions of the azimuth spectrum generator 60B. In the azimuth analysis memory 32c, by extracting reception wave intensity points at the same distance from a plurality of distance spectra generated for each direction, a azimuth spectrum showing the relationship between the reception beam direction and the reception intensity is generated for each distance d.
[0024] Intensity peak extraction program 33f: This is a program for implementing the functions of the peak extraction unit 60C. In the intensity peak analysis memory 32d of the RAM 32, the intensity level and peak position (angle) of the intensity peak are specified for each of the azimuth spectra generated for each distance. The peak position can be determined, for example, by finding three adjacent bins that form a spectrum maximum point, fitting this to a quadratic curve, and then determining the apex position. False image filter processing program 33g: a program for realizing the function of the false image filter unit 60D, which uses the false image filter processing memory 32f of the RAM 32 to select an intensity peak that satisfies the above-mentioned selection condition as a main peak for target identification from a set of intensity peaks extracted on the azimuth spectrum, and then outputs distance and azimuth information indicated by the selected main peak from the output port DO of the input / output unit 34 in Fig. 3 as target detection information.
[0025] The processing flow of each program will be explained below. FIG. 7 shows the process flow of the measurement control program 33a. S1 and S2 are steps carried out by the receiving beam scanning program 33c. First, in S1, beat waveform data X1(t)-Xn(t) from each IC 50M, 50S is taken in, and in S2, a fast Fourier transform (FFT) process is performed to obtain a frequency spectrum, i.e., distance spectrum σ1-σn. Next, in S3, a weighting coefficient w(θ) for the azimuth θ is determined, and in S4, the beat waveform data X1(t)-Xn(t) are multiplied by the weighting coefficient w(θ) and summed to synthesize beat waveform data corresponding to the receiving beam in the azimuth θ. The above process is performed for all quantized azimuth angle positions in the sub-scanning angle range of each IC 50M, 50S.
[0026] Next, the process proceeds to the direction analysis process in S5. FIG. 8 shows the details. In S501, the distance d is set to an initial value d0. In S502, the spectrum intensity value I(d) of the frequency corresponding to the distance d is extracted from the distance spectrum of each direction θ. Then, in S503, the set of (θ, I(d)) is saved as direction spectrum data. In S504, the distance d is incremented by Δd (1 bin), and in S505, the process returns to S502 and repeats the following process until the value of d exceeds the upper limit value dmax.
[0027] Returning to FIG. 7, the peak detection process in S6 follows. FIG. 9 shows the details, and will be described with reference to FIG. 12. The peak detection process is performed on all the azimuth spectrum data obtained for each distance d. First, in S601, the azimuth spectrum data is read. Next, a reference level y' of the azimuth spectrum is set. The reference level y' can be the average or median of the intensity values indicated by each data point of the azimuth spectrum. Next, in S603, spectrum points whose intensity exceeds a threshold value y'+α are extracted as search target points. Next, in S604, spectrum points of three adjacent bins that form the maximum point of the spectrum are searched for. Then, in S605, quadratic curve approximation is performed on the spectrum points that form the maximum points, and in S606, the apex coordinates are specified as the intensity ym and azimuth angle position of the main peak.
[0028] Next, the false image filtering process of S7 in FIG. 7 is executed. The details will be described with reference to the flow chart of FIG. 10 and FIG. 13. Here, the false image peak suppression process is performed by replacing the false image peak in the azimuth spectrum with a lower level intensity peak. First, in S701 in FIG. 10, as shown in FIG. 13, the false image search section Ik is set to [y'+α, ym-β] on the intensity axis. Next, in S702, the false image peak is searched for in the azimuth spectrum by finding three bins that have a maximum value in the false image search section Ik. The false image peak to be suppressed is not subject to the target detection output, and quadratic curve approximation, as performed for the main peak, is not performed. Then, the intensity value of the central one that has a maximum value of the three bins is read as the peak value ys of the false image peak. In S703, the process of uniformly replacing ys with a certain correction intensity level ysc (for example, with y'+α, which is the lower limit value of the false image search section Ik, or a value lower than that) is performed.
[0029] In Figs. 12 and 13, the intensity axis is expressed logarithmically (unit: dB), and the false image peak intensity level Y after reduction is expressed as follows, where y is the intensity level before reduction: Y=α·(y-y') +y' When this is expressed in true values instead of logarithms, the false peak intensity level H after reduction is expressed as follows, where the intensity level before reduction is η and the reference level is η'. H=η α ·η' (1-α) It can be expressed as:
[0030] Returning to FIG. 7, in S8, for each azimuth spectrum for each distance d on which the false image filtering process has been performed, the apex coordinates of each peak are read and compared with a predetermined threshold value for noise removal. Then, in S9, for the peaks exceeding the threshold value, as shown in FIG. 6, the azimuth angle θ indicated by the apex coordinates is associated with the distance d, and the peak is registered as target detection information together with the peak ID (peak identification information) in the target list memory 32g. The target detection information thus registered is output as a detection result in S10 from the output port DO in FIG. 1 or FIG. 3 (if the process has not ended in S11, return to S1 and repeat the following process). FIG. 17 is an example in which the target detection information is mapped and output on a plane spanned by the azimuth angle θ and the distance d. Also, FIG. 16 is a comparative example in which false image peak suppression was not performed. It can be seen that the false images that occurred in large numbers in the comparative example (FIG. 16) can be significantly reduced in the embodiment (FIG. 17).
[0031] In this embodiment, as shown in FIG. 1 or 3, the front end unit 2 for radio wave transmission and reception is composed of a set of a master IC 50M and a slave IC 50S connected by a wiring section SCL, and the frame sampling and data output of the beat waveform signal in the slave IC 50S are synchronously controlled by a synchronization signal from the master IC 50M. According to the inventor's study, it has been found that this type of control causes synchronization deviation between ICs due to internal factors such as variation in characteristic impedance due to the line length of the wiring section SCL and disturbance of the clock signal due to external noise. This synchronization deviation is directly reflected in the phase difference of the beat signal output by each IC, and becomes a cause of false images regardless of the side lobe generated in the response characteristics of the receiving antenna array. Such false images could not be eliminated in principle by conventional side lobe suppression methods, but it can be suppressed without any problems by adopting the present invention.
[0032] In the selection conditions of the above algorithm, an intensity peak that falls outside the upper limit of the false image search range [y'+α, ym-β] is not regarded as a false image and is left as a main peak. In the example shown in Fig. 12 and Fig. 13, there is only one intensity peak greater than ym-β in one azimuth spectrum, and as a result, only one target at a specific distance d from the radio wave transmission source is detected. However, naturally, in this method, if there are multiple intensity peaks that fall outside the upper limit of the false image search range [y'+α, ym-β] in one azimuth spectrum, multiple main peaks are also selected and left. The number of main peaks left by selection varies depending on the setting value of the upper limit adjustment coefficient β set for the intensity value ym of the maximum intensity peak. It can be said that it is desirable to appropriately set the value of β through calibration measurements using various target models, etc., so that the rejection rate of true targets is as small as possible and the rejection rate of false images is as large as possible.
[0033] In this case, by setting the lower limit adjustment coefficient α in the setting of the false image search section to zero, ysc can be made zero, i.e., the false image peak can be completely removed from the azimuth spectrum. On the other hand, by setting the upper limit adjustment coefficient β as small as possible, only the intensity peak with the maximum intensity can be regarded as the main peak, and all other intensity peaks can be discarded as false image peaks, in other words, processing can be performed so that a target object at the same distance from the radio wave transmission source is regarded as only one, and the rest are discarded as false images regardless of their cause.
[0034] The method for reducing (suppressing) the peak value ys of the false image peak is not limited to the method shown in FIG. 12 and FIG. 13, and other methods can be adopted. An example is shown in FIG. 14 and FIG. 15. In this method, as shown in FIG. 14, the protruding difference ys-y' of the false image peak intensity ys from the reference level y' in the logarithmic intensity display is defined as the effective false image peak intensity. Then, as shown in FIG. 15, the value of the effective false image peak intensity (ys-y') / y' standardized by the reference level y' is multiplied by a reduction coefficient γ (where 0<γ<1) to reduce the false image peak intensity ys. According to this method, even if the intensity level of the false image is reduced, the intensity ratio between different false images is preserved, so that the probability of rejecting a true target as a false image during calibration when the reduction coefficient γ is set may be reduced. As described above, the artifact filtering process according to this algorithm is equivalent to replacing S703 in the flow chart of FIG. 10 with S703' and S704' in FIG.
[0035] Figures 18 and 19 show examples of mapping and outputting target detection information by changing the value of γ. Figure 18 shows the case of γ=0.5, and Figure 19 shows the case of γ=0.8, and it can be seen that the number of false images removed changes significantly depending on the value of γ.
[0036] Modified embodiments of the present invention will now be described further below. First, the radar system of the present invention may include a speed identification unit that identifies the speed of a target candidate indicated by an intensity peak extracted by the peak extraction unit in each of the azimuth spectra generated by distance, based on received waveform signals in various azimuths obtained by beam scanning. In this case, the peak selection processing unit may be configured to select, as a main peak, the maximum peak and an intensity peak whose speed is identified as different from that of the maximum peak, among the intensity peaks exceeding a threshold value extracted for each azimuth spectrum. By making it possible to detect targets with different moving speeds as different targets even if they are at the same distance, for example, the probability of erroneously determining that a real target is a counterfeit can be reduced.
[0037] In this case, by modulating the transmission wave for target search with a triangular wave as shown in Fig. 20, it becomes possible to detect the speed of the target using the Doppler shift. This method is well known, and is expressed as follows: Δf: frequency shift width, c: speed of light, d: distance, V: speed, T: frequency shift period, β U : Beat frequency difference (when sweeping up the modulation), β D : Beat frequency difference (during the downward sweep of modulation) d=c·(β U +β D )·T / 4Δf V = c (β U -β D ) / 4Δf As shown in Fig. 21, the speed V is added to the items registered in the target list memory 32g, and even if the distances of targets B3 and B4 are the same (d3 = d4), the former is detected as having a speed of V3 and the latter is detected as having a speed of zero (i.e., stationary), and the intensity peaks of both are selected as main peaks and output as target detection information.
[0038] The measurement control program of Fig. 7 can also be modified as shown in Fig. 22. In this method, the peak extraction unit compares the intensity peaks extracted for each azimuth spectrum with the above-mentioned threshold, and for a set of intensity peaks equal to or greater than the threshold, the information on the distance d, azimuth θ, and intensity value I indicated by each intensity peak is temporarily registered as target candidate information in the target candidate registration unit (target list memory 32g). The peak selection processing unit is further provided with a function of an invalidation processing unit that invalidates registered target candidate information corresponding to intensity peaks that do not satisfy the selection conditions. The target detection output unit then reads out the information on the distance d and azimuth θ included in the remaining target candidate information that has not been invalidated in the target candidate registration unit, and outputs it as target detection information of the main peak.
[0039] Specifically, in FIG. 22, after the peak detection process in S6 is completed, each intensity peak detected in S7' is compared with the above-mentioned threshold, and for intensity peaks equal to or greater than the threshold, the false image filter process is not performed, and the value of the peak intensity I is further incorporated into the target detection information (distance d and direction θ) for all intensity peaks, and a peak ID is assigned and temporarily registered in the target list memory 32g, as shown in the left side of FIG. 23 (FIG. 22: S8'). Then, the false image filter process is performed on this target list memory 32g (FIG. 22: S9'). The process in S9' in FIG. 22 differs from the false image filter process in FIG. 7 in that the data of the intensity peaks determined to be false images is invalidated (for example, deleted as shown in the right side of FIG. 23) so that they are not output as targets. FIG. 23 shows an example in which, among target candidates detected at the same distance, those with small peak intensities (B1 and B2) are invalidated. According to this method, since information on all intensity peaks including false images is stored in the target list memory 32g as target candidate information, it is advantageous in that it becomes easier to process, for example, when it is desired to review the selection conditions for intensity peaks for past measurement data and output the data again.
[0040] 12 to 15, the reference level y' set for the azimuth spectrum is set constant based on, for example, the average value of the spectrum waveform, but the reference level y' may be corrected according to the background swell level. Specifically, the azimuth spectrum is divided into cells (sections) on the azimuth angle axis, and the swell level yud for each cell with respect to the average level y0 is calculated, and the reference level y' for each cell is set as y'=y0+yud According to this method, it is possible to accurately evaluate the false peak level and, ultimately, to select the target object while reducing the influence of waviness. Note that the correction of the peak on the spectrum based on the waviness level can also be performed on the distance spectrum and the velocity spectrum in the same way. Figure 24 shows an example of waviness occurring in the velocity spectrum (beat frequency spectrum).
[0041] The details of the action and effect of the present invention are summarized as follows. As described above in detail, in the present invention, a selection is performed on a plurality of intensity peaks appearing on a distance-specific azimuth spectrum, with the necessary condition being whether the intensity value exceeds a threshold value. Then, distance and azimuth information indicated by the main peak that satisfies the selection condition is output as target detection information. This makes it possible to replace the high-load arithmetic processing that was previously required to suppress the appearance of false images with a simple selection process of intensity peaks. Furthermore, the selection process can effectively suppress intensity peaks resulting from false images that occur due to factors other than jamming waves.
[0042] The peak selection processing unit may be configured to include a false peak suppression unit that identifies an intensity peak in the set of intensity peaks that does not satisfy the selection condition as a false peak and performs a false peak suppression process to delete the false peak in the azimuth spectrum or replace it with an intensity peak of a lower level. In this case, the target detection output unit identifies an intensity peak that exceeds a predetermined threshold in the azimuth spectrum that has been subjected to the false peak suppression process as a main peak, and outputs distance and azimuth information corresponding to the intensity peak as detection information of the target. By lowering or deleting in advance the intensity peaks that do not satisfy the selection condition in the azimuth spectrum, the target detection output unit can identify it as the main peak simply by searching for an intensity peak that exceeds the threshold, thereby further simplifying the processing algorithm.
[0043] On the other hand, a target candidate registration unit may be provided that registers information on the distance, direction, and intensity value of each intensity peak as target candidate information for a set of intensity peaks extracted by the peak extraction unit for each azimuth spectrum. In this case, the peak selection processing unit may be configured to include an invalidation processing unit that invalidates registered target candidate information corresponding to intensity peaks that do not satisfy the selection conditions. The target detection output unit may be configured to read out distance and direction information included in the remaining target candidate information that has not been invalidated by the target candidate registration unit, and output it as target detection information of the main peak. According to this configuration, information on all intensity peaks, including false images, is registered as target candidate information, so that the flexibility of intensity peak data can be improved when it is necessary to respond to changes in selection conditions, and it is also easy to process, for example, when it is necessary to review the selection conditions for intensity peaks for past measurement data and output them again.
[0044] The peak selection processing unit selects the maximum peak with the maximum intensity as the main peak from among the intensity peaks that exceed a threshold value extracted for each distance-based azimuth spectrum. In this method, by assuming that there is only one target at the same distance, it becomes possible to easily identify only the maximum peak in the distance-based azimuth spectrum as the target (main peak). In particular, in the case of stationary targets that exist in different azimuths, it is considered to be an extremely accidental event that their detection distances coincide, and even if a process is performed to regard all targets other than the maximum peak as false images, it is unlikely that this will lead to an erroneous judgment.
[0045] On the other hand, the present invention may include a speed identification unit that identifies the speed of the target candidate indicated by the intensity peak extracted by the peak extraction unit in each of the azimuth spectra generated by distance, based on the received waveform signals in various azimuths obtained by beam scanning. The peak selection processing unit may be configured to select, as the main peak, the maximum peak and an intensity peak whose speed is identified different from that of the maximum peak, among the intensity peaks exceeding a threshold value extracted for each azimuth spectrum. When a moving target is included among multiple targets existing in different azimuths, the distances between the different targets may temporarily become the same. Therefore, by selecting, as the main peak, targets having a different speed from the maximum peak in addition to the target having the maximum peak as described above, it is possible to avoid a problem of erroneously rejecting the latter as a false image.
[0046] The transmitter used in the radar system of the present invention can be configured to transmit, as a search radio wave, a transmission wave whose frequency is continuously modulated according to a predetermined modulation pattern. In this case, a mixer can be provided that mixes the waveform signal of the transmission wave with the waveform signal of the reception wave and outputs a beat waveform signal based on the frequency difference between the transmission wave and the reception wave, and the distance spectrum generator can be configured to generate the frequency spectrum of the beat waveform signal as the distance spectrum. With this method, it is possible to determine the distance to a target with high accuracy using the frequency spectrum of the beat waveform signal.
[0047] In this case, the front-end unit for radio wave transmission and reception of the radar system can be configured to include a master IC and a slave IC cascade-connected to the master IC via a wiring section. A transmitter is incorporated in the master IC, and a plurality of element antennas constituting a receiving antenna array are distributed and incorporated in the master IC and the slave IC. In addition, a mixer that mixes waveform signals of waves received by the element antennas with waveform signals of waves transmitted distributed from the master IC via the wiring section, an A / D converter that digitally converts the beat waveform signal output by the mixer and outputs it serially, and a signal processing device that functions as a target detection output section can be provided. The signal processing device can be provided with an input section to which beat waveform signals from the A / D converters of the master IC and the slave IC are individually input serially, and the slave IC can be provided with a synchronization control section that receives a synchronization signal from the master IC via the wiring section and causes the A / D converter to output a beat waveform signal in synchronization with the master IC in response. The inventors have found through their research that the above configuration is prone to misalignment between ICs due to internal factors such as variations in characteristic impedance caused by wiring length and external noise causing clock signal disturbance. This misalignment is directly reflected in the phase difference of the beat signals output by each IC, and is a cause of false images regardless of the side lobes that occur in the response characteristics of the receiving antenna array. Such false images could not be eliminated in principle by conventional side lobe suppression methods, but the adoption of the present invention makes it possible to suppress them without any problems.
[0048] 1. Radar System 2. Signal Processing Device 3. Front end 31 CPU 32 RAM 33 ROM 33a Measurement control program 33b Digital Filter 33c Receive beam scanning program 33d FFT Module 33e Orientation Analysis Program 33f Intensity peak extraction program 33g False image filtering program 34 Input / output section DO Output Port 30 Bus 41 Transmitting amplifier 42 Phase shifter 43 Low Noise Amplifier 50M Master IC 50S Slave IC SYN0 Synchronous output port SYNI Synchronous input port SCL wiring section WVL Waveform Distribution Line 51 Transmitting Antenna 52 element antenna 53 Mixing section 54 Oscillator Circuit 56 Control circuit (synchronous control section) 55 A / D conversion section 57 Buffer Amplifier 59 Clock Circuit WO1~WOn Waveform output port 60A Distance spectrum generator 60B Azimuth spectrum generator 60C Peak Extraction Unit 60D False Image Filter
Claims
1. a transmitter that transmits a search radio wave as a transmission wave for searching for a target; a receiving section having a receiving antenna array made up of a plurality of element antennas, and receiving a reflected wave of the searching radio wave as a received wave via the receiving antenna array; a scanning unit that performs beam scanning of the received waves by sequentially changing and controlling a receiving beam azimuth of the receiving antenna array within a predetermined azimuth angle range; a distance spectrum generating unit that generates a distance spectrum showing a relationship between distance and reception intensity in the reception beam azimuth for each azimuth based on a transmission waveform signal by the transmitting unit and reception waveform signals in the various reception beam azimuths obtained by the beam scanning; an azimuth spectrum generating unit that generates, for each distance, an azimuth spectrum indicating a relationship between the receiving beam azimuth and the receiving intensity by extracting received wave intensity points at the same distance from the plurality of distance spectra generated for each azimuth; a peak extraction unit that extracts an intensity peak from each of the azimuth spectra generated for each distance; a peak selection processing unit that sets a selection condition for an intensity peak set consisting of the intensity peaks extracted for each of the azimuth spectra, the selection condition including a requirement that an intensity value exceeds a predetermined threshold, and performs a process of selecting an intensity peak from the intensity peak set that satisfies the selection condition as a main peak for target identification; a target detection output unit that outputs information on the distance and direction indicated by the selected main peak as target detection information, The transmission unit transmits, as the search radio wave, a transmission wave whose frequency is continuously modulated in a predetermined modulation pattern, a mixer that mixes the waveform signal of the transmission wave with the waveform signal of the reception wave and outputs a beat waveform signal based on a frequency difference between the transmission wave and the reception wave; the distance spectrum generating unit generates a frequency spectrum of the beat waveform signal as the distance spectrum, a mixing section that mixes a waveform signal of the wave received by the element antenna with a waveform signal of the wave transmitted via a wiring section, and an A / D conversion section that digitally converts the beat waveform signal output by the mixing section and serially outputs the digitalized signal, a signal processing device that functions as the distance spectrum generating unit, the azimuth spectrum generating unit, the peak extracting unit, the peak selecting unit, and the target detection output unit is provided, and the signal processing device is provided with an input unit to which the beat waveform signals from the A / D converting unit are individually and serially input; a synchronization control unit that causes the A / D conversion unit to output the beat waveform signal; a digital filter is provided which takes in beat signal waveform data of the beat waveform signal into a waveform filter memory and removes noise derived from harmonics; the peak selection processing unit includes a false peak suppression unit that performs a false peak suppression process to identify an intensity peak in the intensity peak set that does not satisfy the selection condition as a false peak, and not regard an intensity peak that falls outside an upper limit side of a false peak search interval as the false peak but leave it as the main peak, and to delete the false peak in the azimuth spectrum or replace it with an intensity peak of a lower level; The target detection output unit identifies an intensity peak exceeding a predetermined threshold in the azimuth spectrum on which the false peak suppression process has been performed as the main peak, and outputs distance and azimuth information corresponding to the intensity peak as detection information of the target. A radar system comprising:
2. a target candidate registration unit that registers information on a distance, a direction, and an intensity value indicated by each intensity peak for the intensity peak set extracted by the peak extraction unit for each of the direction spectra as target candidate information; the peak selection processing unit includes an invalidation processing unit that performs invalidation processing on the registered target candidate information corresponding to an intensity peak that does not satisfy the selection condition, 2. The radar system according to claim 1, wherein the target detection output unit reads out distance and direction information contained in the remaining target candidate information that has not been subjected to the invalidation process in the target candidate registration unit, and outputs the distance and direction information as the target detection information of the main peak.
3. 3. The radar system according to claim 1, wherein the peak selection processing unit selects, as the main peak, a maximum peak having a maximum intensity from among the intensity peaks that exceed the threshold and are extracted for each of the azimuth spectra by distance.
4. a speed identification unit that identifies a speed of a target candidate indicated by the intensity peak extracted by the peak extraction unit in each of the direction spectra generated for each distance, based on the received waveform signals in various directions obtained by the beam scanning, 4. The radar system according to claim 3, wherein the peak selection processing unit selects, from the intensity peaks exceeding the threshold value extracted for each of the azimuth spectra, the maximum peak and an intensity peak for which a velocity different from that of the maximum peak is identified as the main peak.
Citation Information
Patent Citations
FM-CW radar device
JP1997133765A
Side lobe canceler device
JP2006162397A
Radar apparatus and target detection method
JP2013200255A
Preceding object recognition device, vehicle control command device, and program and method for recognizing preceding object
JP2014119348A
Radar device and target detection method
JP2019027979A
