JP2015049075A - Radar and object detection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a radar configured to improve an S/N ratio to detect a far object with higher accuracy, and an object detection method.SOLUTION: A transmission unit 21 transmits a transmission signal modulated with a code repeatedly. A speed-detection data acquisition unit 31 repeatedly samples two or more spots of one code of each of reflection waves formed by reflecting the transmission signals on an object. By increasing the samples, an S/N ratio can be improved even if the same noise is included. thereby detecting a smaller signal. A peak search DFT unit 34 and a speed detection unit 35 detect relative speed of the object on the basis of data sampled by the speed-detection data acquisition unit 31, thereby detecting relative speed of a far object with higher accuracy.

Description

  The present invention relates to a radar and an object detection method.

  A direct sequence spread spectrum (DS-SS) type radar has been proposed. As shown in FIG. 15, in the DS-SS radar 1, a transmission signal modulated with a code having a predetermined period of a predetermined sequence (for example, M sequence in FIG. 16) is repeatedly transmitted from the transmission unit 2. The detection data acquisition unit 3 acquires a sampling signal by performing equivalent time sampling with respect to each reflected wave of the transmission signal reflected by the object at a sampling period (N + 1) Tw. As shown in FIG. 17, in the multiplication unit 4, for the sampling signals in the code direction h = 0 to N−1 and the distance direction k = 0 to N−1 after the equivalent time sampling of the sampling period (N + 1) Tw, For each delay time (distance) k, compensation is performed with the same code using a complex conjugate product with the modulation code as a reference function. In the FFT unit 5, Fourier transformation is performed in the code direction for each distance k. Thereby, the speed / distance detection unit 6 obtains the speed (relative speed) V of the object for each distance R.

  For example, in Patent Document 1, means for transmitting a transmission signal including a predetermined code sequence to one or more target bodies, means for receiving a signal reflected by one or more target bodies, and received signals obtained at different times A DS-SS radar having means for calculating the sum and difference of the signals and calculating the Doppler frequency shift of the target object from the phase difference between the sum signal and the difference signal is disclosed.

JP 2008-232668 A

  However, in the above technique, since the S / N ratio of the detection signal depends on the code length, when a relatively long code cannot be used, there is a problem that the S / N ratio is insufficient and the detection distance is short. Improvement is desired.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a radar and an object detection method that can improve an S / N ratio and detect a distant object with higher accuracy.

  The present invention relates to a transmission unit that repeatedly transmits a transmission signal modulated with a code, a speed detection data acquisition unit that repeatedly samples two or more portions of one code of a reflected wave of each transmission signal reflected by an object, and a speed Based on the data sampled by the detection data acquisition unit, a speed detection unit that detects the relative speed of the object, and distance detection data acquisition that samples the reflected wave of each transmission signal reflected by the object at a sampling period that can restore the code Based on the relative velocity of the object detected by the velocity detection unit, the compensation unit for compensating the Doppler shift of the reflected wave sampled by the distance detection data acquisition unit, and the reflected wave compensated for the Doppler shift by the compensation unit The radar includes a distance detection unit that detects a distance to an object based on the distance.

  According to this configuration, the transmission unit repeatedly transmits a transmission signal modulated with a code. The speed detection data acquisition unit repeatedly samples two or more portions of one code of the reflected wave of each transmission signal reflected by the object. By increasing the number of samples in this way, the S / N ratio can be improved and a smaller signal can be detected even if the same noise is included. For this reason, the speed detection unit detects the relative speed of the object based on the data sampled by the speed detection data acquisition unit, so that the relative speed of the farther object can be detected with higher accuracy.

  Further, according to this configuration, the distance detection data acquisition unit samples each reflected wave of the transmission signal reflected by the object at a sampling period at which the code can be restored. The compensation unit compensates for the Doppler shift of the reflected wave sampled by the distance detection data acquisition unit based on the relative velocity of the object detected by the velocity detection unit. Thereby, the Doppler compensation of the reflected wave can be performed based on the relative velocity detected with higher accuracy. The distance detection unit detects the distance from the object based on the reflected wave that has been compensated for Doppler shift by the compensation unit. As a result, the distance to the object can be detected based on the reflected wave that has been Doppler compensated at the relative velocity detected with higher accuracy.

  In this case, the velocity detection unit detects the relative velocity of the object based on the non-coherent addition of the Fourier transform of each set of data at the same position of the reflected wave sampled by the velocity detection data acquisition unit. It is preferable to do.

  According to this configuration, the speed detection unit is configured to perform relative comparison of objects based on a non-coherent addition of a Fourier transform of each set of data at the same position of the reflected wave code sampled by the speed detection data acquisition unit. Detect speed. As a result, it is possible to obtain the true peak of the signal by frequency decomposition.

  The present invention also includes a transmission step of repeatedly transmitting a transmission signal modulated with a code, a speed detection data acquisition step of repeatedly sampling two or more locations with one code of each reflected wave of the transmission signal reflected by an object, and Based on the data sampled in the speed detection data acquisition step, the speed detection step for detecting the relative speed of the object, and the distance detection data for sampling the reflected wave of the transmission signal reflected on the object at a sampling period capable of restoring the code An acquisition step, a compensation step for compensating the Doppler shift of the reflected wave sampled in the distance detection data acquisition step based on the relative velocity of the object detected in the velocity detection step, and a reflected wave in which the Doppler shift is compensated by the compensation step And a distance detection step of detecting a distance to the object based on the above.

  In this case, in the speed detection process, the relative speed of the object is detected based on the non-coherent addition of the Fourier transform of each set of data at the same position of the sign of the reflected wave sampled in the speed detection data acquisition process. It is preferable to do.

  According to the radar and the object detection method of the present invention, the S / N ratio can be improved and a farther object can be detected with higher accuracy.

It is a block diagram which shows the DS-SS radar of embodiment. It is a figure which shows the signal processing of the DS-SS radar of embodiment. It is a graph which shows the estimation result of the speed before and behind the non-coherent addition of the sample direction of embodiment. It is a graph which shows the estimation result of the distance of embodiment. It is a table | surface which shows the calculation result of the maximum detection distance of the vehicle in the short distance mode of the DS-SS radar of embodiment. It is a table | surface which shows the calculation result of the human maximum detection distance in the short distance mode of the DS-SS radar of embodiment. It is a table | surface which shows the calculation result of the maximum detection distance of the vehicle in the long-distance mode of the DS-SS radar of embodiment. It is a table | surface which shows the calculation result of the human maximum detection distance in the long-distance mode of the DS-SS radar of embodiment. It is a table | surface which shows the parameter of the short-distance mode of a radar, and the expected performance in the simulation of the DS-SS radar of embodiment. It is a table | surface which shows the simulation conditions of the DS-SS radar of embodiment. (A) and (b) are conditions No. of FIG. 2 is a graph showing the results of distance estimation of 1 simulation, where (a) shows the result of target 1 and (b) shows the result of target 2. FIG. (A) and (b) are conditions No. of FIG. 2 is a graph showing the results of distance estimation of the simulation of FIG. 2, (a) shows the result of target 1, and (b) shows the result of target 2. (A) and (b) are conditions No. of FIG. 3 is a graph showing the results of distance estimation of the simulation of FIG. 3, (a) shows the result of target 1, and (b) shows the result of target 2. (A) and (b) are conditions No. of FIG. 4A and 4B are graphs showing the results of distance estimation of the simulation of FIG. 4, in which FIG. It is a block diagram which shows the conventional DS-SS radar. It is a figure which shows the signal transmitted with DS-SS radar. It is a figure which shows the signal processing of the conventional DS-SS radar. (A)-(d) is a graph which shows the signal processing result about one target without noise, (a) (b) shows a speed characteristic, (c) (d) shows a distance characteristic. It is a graph which shows the frequency spectrum of the transmission signal in code length N = 127, and the received signal after a Doppler shift. It is a table | surface which shows the calculation result of the maximum detection distance in the radar reflection cross section 15dBm of the vehicle of the conventional DS-SS radar. It is a table | surface which shows the calculation result of the maximum detection distance in the radar reflection cross section 10dBmm of the vehicle of the conventional DS-SS radar. It is a table | surface which shows the parameter of the short-distance mode and the expected performance of the radar in the simulation of the conventional DS-SS radar. It is a table | surface which shows the conditions and result of the simulation of the conventional DS-SS radar. Condition No. in FIG. 2 is a graph showing the results of 1 simulation. (A)-(d) shows condition No. of FIG. (A) shows the speed bin for target 1; (b) shows the distance bin for target 1; (c) shows the speed bin for target 2; (D) shows the distance bin for target 2. It is a graph which shows the reception power before A / D sampling.

  An example of a radar and an object detection method according to an embodiment of the present invention will be described with reference to the drawings. As shown in FIG. 1, the DS-SS radar 10 according to the embodiment of the present invention includes a transmission unit 21, a speed detection data acquisition unit 31, an FFT unit 32, a sample direction non-coherent addition unit 33, a peak search DFT unit 34, A speed detection unit 35, a distance detection data acquisition unit 41, a Doppler compensation unit 42, a multiplication unit 43, a code direction coherent addition unit 44, and a distance detection unit 45 are provided.

  As shown in FIG. 16, the transmission unit 21 continuously and repeatedly transmits a signal whose phase is code-modulated with an M-sequence code for each chip. The chip width of one chip of the signal is Tw, the code length of codes n = 0 to N−1 included in the signal is N, and the period m = 0 to M−1 with respect to the observation time Tc required for one signal processing Let M be the number.

As shown in FIG. 2, the speed detection data acquisition unit 31 responds to the reflected wave of the transmission signal reflected by the object, as shown in FIG. Equivalent time sampling is performed on the received signal in the interval of (1), with sp interval samples (where the code length is smaller than N, and sp is preferably a power of 2). As a result, sp pieces of sampling data having the same code at the sampling interval N are obtained. That is, in this embodiment, in order to improve the S / N, a plurality of sampling data with the same code at the sampling interval N is obtained by repeatedly sampling two or more locations of one code of the reflected wave. With respect to one code, sampling data of at least 2 chips is obtained, and sampling data of all chips is obtained at most. Here, since the code length of the M-sequence code is not a power of 2 ⁿ −1 and 2, when performing equivalent time sampling, it is preferable to delay the sampling acquisition timing by one chip for each period.

  As shown in FIG. 2, the FFT unit 32 performs discrete Fourier transform in the direction of the number of periods m with respect to the sampling data for the same code after the equivalent time sampling, and obtains speed estimation results for the number of samples.

  As shown in FIG. 2, the sample direction non-coherent addition unit 33 performs addition of each power, that is, non-coherent addition, on the speed estimation result for the number of samples after the discrete Fourier transform in the period m direction. FIG. 3 shows simulation speed estimation results at a distance R = 0 m, a relative speed V = 30 km / h, and an S / N ratio = 0 dB. As shown in FIG. 3, it can be seen that the non-coherent addition in the sample direction suppresses the dispersion of noise in the frequency domain. As a result, the S / Nmin, which is the minimum S / N value at the time of target detection, can be lowered, and further extension of the detection distance is expected.

  The peak search DFT unit 34 calculates the Doppler frequency peak frequency number detected from the velocity estimation result after non-coherent addition in the sample direction in order to reduce the estimation error of the relative velocity of the object when performing Doppler compensation in the distance estimation. The relative velocity of the object is estimated by discrete Fourier transform for the front and rear 1 bin range.

  The velocity detection unit 35 outputs the relative velocity of the object estimated by the discrete Fourier transform and transmits it to the Doppler compensation unit 42.

  As shown in FIG. 2, the distance detection data acquisition unit 41 performs equivalent time sampling with N + 1 interval samples on the received signal in the interval of the distance estimation interval observation time Tc_R and the period Mr (= N), Get direction sampling data.

  The reception signal of equivalent time sampling is influenced by the Doppler shift, and the phase becomes discontinuous in the time direction. Therefore, as shown in FIG. 2, the Doppler compensation unit 42 performs Doppler compensation using information on the target speed obtained by the speed detection unit 35.

  As shown in FIG. 2, the multiplication unit 43 performs code compensation by multiplying the received signal after Doppler compensation by an M-sequence code that is a reference function of the transmission signal shifted by one distance from the distance delay.

  As shown in FIG. 2, the code direction coherent addition unit 44 performs coherent addition of the number of samples Mr in the code direction on the received signal after multiplication. FIG. 4 shows the distance estimation results of the simulation when the distance R = 10 m, the relative speed V = 30 km / h, the speed estimation error Error = 0, and no noise. As shown in FIG. 4, since the phase is continuous in the distance of the object, the estimation result of the distance of the object can be obtained by coherent addition in the code direction. It can be seen that under the simulation condition under one target, the side lobe level is 20 log (2047) = about 62 dB based on the code length N = 2047. The distance detection unit 45 outputs the distance of the object obtained by coherent addition in the code direction.

  According to the present embodiment, the transmission unit 21 repeatedly transmits a transmission signal modulated with a code. The speed detection data acquisition unit 31 repeatedly samples two or more portions of one code of the reflected wave of each transmission signal reflected by the object. By increasing the number of samples in this way, the S / N ratio can be improved and a smaller signal can be detected even if the same noise is included. For this reason, the peak search DFT unit 34 and the velocity detection unit 35 detect the relative velocity of the object based on the data sampled by the velocity detection data acquisition unit 31, so that the relative velocity of a distant object is more accurately detected. Can be detected.

  Further, according to the present embodiment, the distance detection data acquisition unit 41 samples the reflected wave of each transmission signal reflected by the object at a sampling period at which the code can be restored. The Doppler compensation unit 42 compensates for the Doppler shift of the reflected wave sampled by the distance detection data acquisition unit 41 based on the relative velocity of the object detected by the velocity detection unit 35. Thereby, the Doppler compensation of the reflected wave can be performed based on the relative velocity detected with higher accuracy. The code direction coherent adder 44 and the distance detector 45 detect the distance to the object based on the reflected wave that has been compensated for the Doppler shift by the Doppler compensator 42. As a result, the distance to the object can be detected based on the reflected wave that has been Doppler compensated at the relative velocity detected with higher accuracy.

  In addition, according to the present embodiment, the peak search DFT unit 34 and the velocity detection unit 35 perform a non-linear transformation on a set of data at the same position of the sign of the reflected wave sampled by the velocity detection data acquisition unit 31. The relative velocity of the object is detected based on the coherent addition. As a result, it is possible to obtain the true peak of the signal by frequency decomposition.

  Hereinafter, the simulation result about the DS-SS radar 10 of this embodiment shown in FIG. 1 is shown. It should be noted that during the speed estimation observation time Tc = Tw · N · Mv = 1 nsec · 2047 chip · 2048 period = about 4 msec, the range walk amount at which the target of the relative speed Vtg = 200 km / h moves is Vtg · Tc = 0. 22m. Therefore, the range walk amount in the section of the observation time of 32 msec is 1.78 m, and signal loss due to the range walk corresponding to about 10 times the chip width of 1 msec (distance resolution 0.15 m) occurs.

  Therefore, for the short distance target, a short distance mode is used in which the observation time is 4 msec and non-coherent addition is performed in the sample direction. For long-distance targets, by increasing the chip width to 16 nsec (bandwidth 62.5 MHz, distance resolution 2.4 m), (1) reducing the receiver bandwidth, reducing the noise (kTB product) in the receiver, and (2) The long-distance mode is used in which the influence of the range walk is mitigated and the signal processing gain is improved by extending the observation time for speed estimation. The signal processing was switched in a time-sharing manner for both the short-distance mode and the long-distance mode.

  Calculation results of the maximum detection distance in the short-distance mode and the long-distance mode in the DS-SS radar 10 of the present embodiment are shown in FIGS. 5 shows the maximum detection distance of the vehicle in the short distance mode, FIG. 6 shows the maximum detection distance of the human in the short distance mode, FIG. 7 shows the maximum detection distance of the vehicle in the long distance mode, and FIG. Indicates the maximum human detection distance in the mode. It can be seen that the maximum detection distance is improved as compared with the conventional DS-SS radar 1 described later.

  In the short distance mode of the DS-SS radar 10 of the present embodiment, a simulation for detecting two objects was performed. Radar parameters and expected performance are shown in FIG. 9, and simulation conditions are shown in FIG. 11A and 11B, the simulation condition No. The distance estimation results of speeds 30 km / h and 40 km / h (detection peak numbers 18 and 25) detected in 1 are shown. In both cases, the side lobe (floor) level was about 32 dB. This is probably because Doppler correction processing is performed on the signal in the distance estimation section using each peak information, and therefore another target Doppler component remains during distance estimation processing for each Doppler. Simulation condition no. No. 1 is an environment where two Doppler targets having different signal levels of S1 / S2 = 0 dB exist, and there is no noise of the conventional DS-SS radar 1 of the later-described type. (Floor) level.

  Simulation condition no. 2-No. 4, the signal levels S1 / S2 of the two objects were changed in increments of 10 dB. Simulation condition no. 2-No. The distance estimation results of 4 are shown in FIGS. 12 (a), 12 (b) to 14 (a), (b), respectively. As shown in FIGS. 12A and 12B to FIGS. As expected from the result of 1, the side lobe (floor) level is decreased in the target 1 having a signal level larger than that of the target 2, and the side lobe (floor) level is increased in the target 2 on the contrary, as shown in FIG. b) to simulation conditions No. 1 shown in FIGS. 3 and no. In case of 4, it was difficult to detect the distance. However, as will be described later, the simulation condition no. Under the condition of 4, the conventional DS-SS radar 1 was difficult to detect the target. Therefore, it seems that the DS-SS radar 10 of this embodiment has improved capability with respect to the target detection capability.

  For comparison, simulation results for the conventional DS-SS radar 1 shown in FIG. 15 are shown. First, FIG. 18A to FIG. 18D show the processing results under the condition that one target object without noise exists. The code length N = 127. As shown in FIGS. 18A to 18D, the floor is lowered in the speed bin and the distance bin corresponding to the target distance / speed, and the side lobe level is high in the distance / speed where the target does not exist. It is confirmed that 10 log (127) ≈21 dB depending on the length. As these factors, since the phase after multiplication becomes discontinuous at a distance where the target does not exist, it is conceivable that the target Doppler component is diffused as noise in the distance where the target does not exist.

  FIG. 19 shows the result of discrete Fourier transform of the transmission signal at the code length N = 127 and the reception signal after Doppler shift. FIG. 19 shows that the transmission signal has a filter characteristic that lowers only frequency number 0, and the frequency spectrum after Doppler shift of frequency number 10 lowers only frequency number 10. From this, it can be considered that the target Doppler component diffuses as noise even at a speed where the target does not exist. The S / N improvement capability in this method depends on the S / N improvement gain by FFT, and is expected to be 10 log (N) (about 33 dB when the code length N = 2047).

  Next, FIG. 20 and FIG. 21 show the calculation results of the maximum detection distances of the automobile and the human in the conventional DS-SS radar 1. Here, the receiver bandwidth is 1.5 GHz which is 1.5 times the bandwidth of one chip, the gains of the transmission antenna and the reception antenna are 23 dB, and the target detection threshold is S / Nmin = 13 dB. As shown in FIGS. 20 and 21, it can be seen that the maximum detection distances of automobiles and humans in the conventional DS-SS radar 1 are inferior to the DS-SS radar 10 of the present embodiment described above.

  Next, a simulation for detecting two objects was performed in the conventional DS-SS radar 1. Radar parameters and expected performance are shown in FIG. 22, and simulation conditions are shown in FIG. In the simulation of the conventional DS-SS radar 1, two targets are arranged at a distance of 10 m and a distance of 100 m, respectively, and simulation was performed with and without consideration of distance attenuation (−40 dB). . The noise was SNR = 0 dB, −10 dB, and −20 dB. FIG. 23 shows the results of S / N improvement under each simulation condition. The threshold for target detection was set to S / Nmin = 13 dB, which is the minimum S / N for target detection, and an object exceeding this threshold was detected. For reference, simulation condition no. The results of 5 are shown in FIGS. 24 and 25 (a) to 25 (d). From FIG. 23, it is difficult for the DS-SS radar 10 of the present embodiment to detect the distance, but the target detection is performed under a condition that allows for possible distance attenuation (−40 dB). It turns out that even target detection is difficult.

  The received power before A / D sampling in the conventional DS-SS radar 1 is shown in FIG. At this time, the gain of the transmission / reception antenna, the radar reflection cross-sectional area RCS between the vehicle and the human, and the signal processing gain were the same values as in the maximum detection distance calculation in the conventional DS-SS radar 1 described above. In addition, the isolation of transmission wave leakage was set to 50 dB. FIG. 26 shows that the target signal is buried below the noise at a distance corresponding to the intersection or less of the reflected power of the automobile or human and the noise power after the signal processing. Further, since the side lobe level of the conventional DS-SS radar 1 is 33 dB, a weak signal may be buried in the target side lobe of a strong signal in a multi-target environment with a target power difference larger than this. Conceivable.

  In addition, this invention is not limited to the said embodiment, A various deformation | transformation aspect is possible.

  DESCRIPTION OF SYMBOLS 1 ... DS-SS radar, 2 ... Transmission part, 3 ... Data acquisition part for detection, 4 ... Multiplication part, 5 ... FFT part, 6 ... Speed / distance detection part, 10 ... DS-SS radar, 21 ... Transmission part, 31 ... Speed detection data acquisition unit, 32 ... FFT unit, 33 ... Sample direction non-coherent addition unit, 34 ... Peak search DFT unit, 35 ... Speed detection unit, 41 ... Distance detection data acquisition unit, 42 ... Doppler compensation unit , 43... Multiplication unit, 44... Code direction coherent addition unit, 45.

Claims (4)

A transmitter that repeatedly transmits a transmission signal modulated by a code;
A speed detection data acquisition unit that repeatedly samples two or more locations of one of the reflected waves of each of the transmission signals reflected by an object;
Based on the data sampled by the speed detection data acquisition unit, a speed detection unit that detects a relative speed of the object;
A distance detection data acquisition unit that samples the reflected wave of each of the transmission signals reflected by the object at a sampling period capable of restoring the code;
A compensation unit that compensates for a Doppler shift of the reflected wave sampled by the data detection unit for distance detection based on the relative velocity of the object detected by the velocity detection unit;
A radar comprising: a distance detection unit that detects a distance to the object based on the reflected wave that has been compensated for Doppler shift by the compensation unit.   The speed detection unit is based on a non-coherent addition of a set of data of the reflected wave sampled by the speed detection data acquisition unit at the same position of the sign and Fourier-transformed. The radar according to claim 1, wherein the radar is detected. A transmission step of repeatedly transmitting a transmission signal modulated with a code;
A speed detection data acquisition step of repeatedly sampling two or more locations with one code of the reflected wave of each of the transmission signals reflected on an object;
A speed detection step of detecting a relative speed of the object based on the data sampled in the speed detection data acquisition step;
A distance detection data acquisition step for sampling a reflected wave of the transmission signal reflected by the object at a sampling period that can restore the code;
A compensation step for compensating for a Doppler shift of the reflected wave sampled in the distance detection data acquisition step based on the relative velocity of the object detected in the velocity detection step;
A distance detection step of detecting a distance from the object based on the reflected wave that has been compensated for Doppler shift by the compensation step;   In the velocity detection step, the relative velocity of the object is based on a non-coherent addition of a set of data of the reflected wave sampled in the velocity detection data acquisition step at the same position with the same sign and Fourier transform. The object detection method according to claim 3, wherein the object is detected.