Radar device

Abstract

The present invention provides a radar device (100), comprising: a data calculation unit (10), a first threshold calculation unit (10), a second threshold calculation unit (10), a detection threshold calculation unit (10), and a peak detection unit (10). The data calculation unit performs frequency analysis on a received signal to calculate complex data. The first threshold calculation unit calculates a first threshold by adding a power value of average data obtained by averaging the complex data to an addition value. The second threshold calculation unit calculates a second threshold based on noise power. The detection threshold calculation unit calculates the larger value of the first threshold and the second threshold according to the frequency band as the detection threshold.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This international application claims the benefit of priority based on Japanese Patent Application No. 2018-092390 filed with the Japan Patent Office on May 11, 2018, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to techniques for detecting peaks from received signals. Background Art

[0004] The radar device installed in the bumper of the vehicle transmits and receives radar waves and obtains received signals, and detects other vehicles, pedestrians and other targets around the vehicle based on the received signals. However, the received signals obtained by the radar device installed in the vehicle may include internal reflection signals of the vehicle in addition to the reflection signals from the target. The internal reflection signals of the vehicle include reflection signals from the bumper or antenna cover, reflection signals generated in the internal parts of the vehicle and transmitted inside the vehicle, etc.

[0005] Generally, when detecting a target by a radar device, the noise power value of the radar device is estimated in advance, and the value obtained by adding a predetermined power value to the estimated noise power value is set as the detection threshold. Then, the peak of the spectrum power of the received signal that is greater than the detection threshold is detected. When the same target detection method is applied to a radar device mounted on a vehicle, there may be a case where the power value of the internal reflection signal that is not originally intended to be detected exceeds the detection threshold, and not only the peak of the reflection signal from the target is detected, but also the peak of the internal reflection signal is detected. However, since the internal reflection signal of the vehicle changes in a complex manner depending on the mounting position of the radar device and the relative position of the radar device and the internal parts of the vehicle, it is difficult to estimate in advance what kind of internal reflection signal is received. Therefore, even if the peak of the internal reflection signal is detected, it is difficult to distinguish the peak of the internal reflection signal from the peak of the reflection signal from the target present near the radar device.

[0006] Therefore, as a method for removing static clutter components from a received signal received by a radar device, a Moving Target Indication (MTI) process has been proposed. The MTI process generates a received signal in which only a signal with static clutter components without amplitude phase fluctuations is suppressed by applying a high-pass filter to each frequency bin of the power spectrum of the received signal in a time series. Furthermore, the MTI process detects a detection threshold for the generated received signal and compares the power of the received signal with the detection threshold to detect a peak.

[0007] Non-patent literature 1: Mark A. Richards, Fundamentals of Radar Signal Processing, Chapter 5: Doppler Processing

[0008] However, MTI processing cannot completely suppress a signal if the amplitude or phase component of the signal to be suppressed does not completely stop in time or if the radar system characteristics vary in time. Therefore, if the amplitude or phase component of the signal to be suppressed varies slightly, there is a possibility that the power of the signal to be suppressed is greater than the detection threshold and is detected as a peak. As a result of detailed research by the inventors, it was found that when MTI processing is applied to a received signal acquired by a radar device mounted on a vehicle, the amplitude or phase component of the internal reflection signal varies slightly in time, and the internal reflection signal cannot be fully suppressed, resulting in a possible problem of erroneous detection of the internal reflection signal. Summary of the invention

[0009] The present disclosure is intended to suppress erroneous detection of an internal reflection signal of the vehicle in a radar device mounted on the vehicle.

[0010] One aspect of the present disclosure is a radar device mounted on a vehicle, comprising: a transmitting antenna, a receiving antenna, a signal acquisition unit, a data calculation unit, an average calculation unit, a first threshold calculation unit, a second threshold calculation unit, a detection threshold calculation unit, and a peak detection unit. The transmitting antenna is configured to transmit radar waves. The receiving antenna is configured to receive reflected waves generated by the reflection of radar waves. The signal acquisition unit is configured to acquire a received signal based on the reflected waves. The data calculation unit is configured to perform frequency analysis on the received signal acquired by the signal acquisition unit to calculate complex data as a frequency function. The average calculation unit is configured to calculate average data. The average data is data obtained by averaging multiple complex data in a specified period before the current processing cycle according to the frequency band. The first threshold calculation unit is configured to add a specified addition value to the power value of the average data calculated by the average calculation unit to calculate the first threshold as a frequency function. The second threshold calculation unit is configured to calculate the second threshold as a frequency function based on the noise power in the observation performed by the radar device. The detection threshold calculation unit is configured to compare the first threshold calculated by the first threshold calculation unit and the second threshold calculated by the second threshold calculation unit according to the frequency band, and calculate the larger value as the detection threshold. The peak detection unit is configured to compare the power value of the complex data in this processing cycle with the detection threshold according to the frequency band, and detect the frequency at which the power value of the complex data is greater than the detection threshold and becomes maximum as the peak.

[0011] According to one aspect of the present disclosure, the average data is calculated by averaging the complex data calculated in a specified period by frequency band. Moreover, the power value of the average data is added with an addition value to calculate the first threshold. Since the relative position of the vehicle's bumper, internal parts, etc. and the radar device is almost unchanged during the specified period, the phase of the amplitude of the complex data representing the vehicle's internal reflection signal in each processing cycle is almost unchanged. On the other hand, during the specified period, since the relative position of the target and the radar device changes, the amplitude and / or phase of the complex data representing the target reflection signal in each processing cycle changes randomly. Therefore, the power value of the average data is almost unchanged in the frequency band where the internal reflection signal is observed, and is relatively greatly reduced in the frequency band where the target reflection signal is observed, compared with the power value of the complex data of this processing cycle. Therefore, the influence of the target reflection signal on the first threshold is relatively small, and the first threshold is a value of the same degree as the value obtained by adding the addition value to the power value of the complex data representing the internal reflection signal.

[0012] Furthermore, the first threshold value and the second threshold value calculated based on the noise power are compared, and the larger value is calculated as the detection threshold value. That is, when the power value of the internal reflection signal is larger than the noise power, the first threshold value based on the power value of the internal reflection signal is calculated as the detection threshold value. Furthermore, the power value of the complex data in this processing cycle is compared with the detection threshold value to detect the frequency of the peak value. Therefore, by feeding back the power value of the average data to the detection threshold value, it is possible to suppress the false detection of the internal reflection signal of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a diagram showing the configuration of a radar device according to the first embodiment.

[0014] Figure 2 It is a diagram showing an FMCW modulation waveform.

[0015] Figure 3 : is a flowchart showing the procedure of the target information acquisition process according to the first embodiment.

[0016] Figure 4 It is a diagram showing a timing chart of a beat signal.

[0017] Figure 5 : is a diagram showing the frequency spectrum of the beat signal.

[0018] Figure 6 This is a diagram for explaining reflection from a target and reflection inside a vehicle.

[0019] Figure 7 This is a diagram showing the spectrum power of target reflection and vehicle interior reflection and the threshold value obtained from the noise power.

[0020] Figure 8 This is a diagram for explaining the peak detection method in the MTI process.

[0021] Fig. 9 This is a flowchart showing the procedure of the peak detection process according to the first embodiment.

[0022] Fig.10 This is a diagram showing vectors of average data when the relative position between the radar device and the reflector changes.

[0023] Fig.11 This is a diagram showing vectors of average data when the relative position between the radar device and the reflector does not change.

[0024] Fig.12 It is a diagram showing power spectra of target reflection and vehicle interior reflection, a threshold value obtained from noise power, and a detection threshold value according to the first embodiment.

[0025] Fig.13 It is a flowchart showing the procedure of the peak detection process involved in the second embodiment.

[0026] Fig.14 It is a diagram showing power spectra of target reflection and vehicle interior reflection, and detection thresholds according to the second embodiment. DETAILED DESCRIPTION

[0027] Hereinafter, a mode for implementing the present disclosure will be described with reference to the drawings.

[0028] (First Embodiment)

[0029] <1. Structure>

[0030] First, refer to Figure 1 The structure of the radar device 100 according to the present embodiment will be described. The radar device 100 is a Frequency Modulated Continuous Wave (FMCW) radar device. Figure 6 As shown, the radar device 100 is mounted on the center of the vehicle width direction inside the front bumper 81 of the vehicle 80 .

[0031] The radar device 100 includes a processing device 10 , a tilt wave generator 20 , a transmission antenna 30 , a reception antenna 40 for K (K is a natural number) channels, K mixers 50 , and an AD converter 60 .

[0032] The processing device 10 is mainly composed of a microcomputer including a CPU 11, a ROM 12, a RAM 13, and an I / O. The processing device 10 generates a frequency control signal for setting the frequency of a transmission signal, and transmits the generated frequency control signal to the tilt wave generator 20. The tilt wave generator 20 generates a radar signal based on the frequency control signal received from the processing device 10, and transmits the generated radar signal to the transmission antenna 30. In addition, the tilt wave generator 20 supplies the generated radar signals to the K mixers 50, respectively.

[0033] The transmitting antenna 30 radiates radar waves that have been FMCW-modulated based on the radar signal received from the tilt wave generator 20. Specifically, Figure 2 As shown, the transmitting antenna 30 repeatedly transmits a combination of an uplink tilt wave and a downlink tilt wave in each processing cycle, wherein the frequency of the uplink tilt wave monotonically increases from the carrier frequency fc to fc+F within the modulation time T, and the frequency of the downlink tilt wave monotonically decreases from fc+F to fc within the modulation time T.

[0034] The K receiving antennas 40 are arranged in a row in the horizontal direction. Each receiving antenna 40 receives a reflected wave generated when a radar wave is reflected by a reflector, and supplies the reflected signal to the mixer 50 .

[0035] K mixers 50 are provided one for each receiving antenna 40. Each mixer 50 mixes the radar signal supplied from the tilt wave generator 20 and the reflected signal supplied from the receiving antenna 40, and generates a frequency difference signal (hereinafter referred to as a beat signal) B(t) having the difference in frequency between the radar signal and the reflected signal as a frequency component. Figure 4 The timing diagram of the beat signal B(t) is shown in FIG. The beat signal B(t) is represented by the following equation (1). Fb represents the beat frequency, the value obtained by adding the two items represents the beat frequency during the downward tilt, and the value obtained by subtracting the latter item from the former item represents the beat frequency during the upward tilt. θi represents the initial phase. In addition, c represents the speed of light, v represents the speed of the target relative to the radar device 100, and r represents the distance from the radar device 100 to the target.

[0036] [Formula 1]

[0037]

[0038] Then, each mixer 50 transmits the generated beat signal B(t) to the AD converter 60. In the present embodiment, the beat signal B(t) corresponds to the received signal, and the mixer 50 and the AD converter 60 correspond to the signal acquisition unit.

[0039] The AD converter 60 samples the beat signals B(t) of the K channels sent from the K mixers 50, generates discrete beat signals b(t), and sends the generated discrete beat signals b(t) of the K channels to the processing device 10. Specifically, the AD converter 60 is clock-synchronized with the ramp wave generator 20, and in each processing cycle, starts sampling the beat signal B(t) after a predetermined time offset from the start of radar wave transmission, and samples a predetermined time interval.

[0040] The processing device 10 performs signal processing such as frequency analysis on the discrete beat signal of the K channel obtained from the AD converter 60. The processing device 10 loads the program code stored in the non-transitional physical recording medium such as ROM12 through the CPU 11 and executes the program code to realize the functions of the data calculation unit, the average calculation unit, the first threshold calculation unit, the second threshold calculation unit, the detection threshold calculation unit, and the peak detection unit. The method of realizing these functions is not limited to software, and hardware such as combinational logic circuits and analog circuits can also be used to realize part or all of the functions. In addition, the detection signal from the speed sensor that detects the speed of the vehicle 80 is input to the processing device 10.

[0041] <2. Processing>

[0042] <2-1. Overall treatment>

[0043] Next, refer to Figure 3 The procedure of the target information acquisition process executed by the processing device 10 will be described with reference to the flowchart of FIG. 1. The processing device 10 repeatedly executes the target information acquisition process every time a radar wave is transmitted.

[0044] First, in S10 , the processing device 10 acquires a discrete beat signal b(t) of an upward ramp wave and a downward ramp wave of the K channel sampled by the AD converter 60 .

[0045] Next, in S20, the processing device 10 performs complex FFT processing on the discrete beat signal b(t) during uplink tilt and downlink tilt respectively according to the channel, such as Figure 5 As shown, the FFT complex data is calculated for the uplink tilt and the downlink tilt respectively. Complex FFT processing is an example of frequency analysis. As represented by the following formula (2), the FFT complex data S (t0, fb) at the moment t0 of this processing cycle is a function of the beat frequency fb. The FFT complex data S (t0, fb) contains information about the amplitude and its phase. Even if the reflector moves slightly, the phase of the FFT complex data S (t0, fb) reacts sensitively. For example, when the carrier frequency fc is 77 GHz, if the position of the reflector changes by 1 mm, the phase rotates by 185°.

[0046] [Formula 2]

[0047]

[0048] The power spectrum Ps of the FFT complex data S(t0,fb) is a combination of Figure 7 The solid line and the dashed line graph. The beat frequency fb is related to the distance r and represents the distance information. The processing device 10 adds the FFT complex data S(t0, fb) of the K channels during the uplink tilt according to the frequency band to calculate the synthetic complex data Sk(t0, fb) during the uplink tilt. In addition, the processing device 10 adds the FFT complex data S(t0, fb) of the K channels during the downlink tilt according to the frequency band to calculate the synthetic complex data Sk(t0, fb) during the downlink tilt. In this embodiment, the synthetic complex data Sk(t0, fb) is equivalent to the complex data.

[0049] Next, in S30, the processing device 10 compares the synthesized complex data Sk (t0, fb) during the up-tilt and down-tilt with the detection threshold described later, respectively, by frequency band. Furthermore, the processing device 10 detects the beat frequency fb (hereinafter referred to as the peak frequency) at which the power value of the synthesized complex data Sk (t0, fb) is greater than the detection threshold and becomes maximum from the synthesized complex data Sk (t0, fb) during the up-tilt and down-tilt, respectively, as the peak. When observing multiple targets, multiple peak frequencies are detected. In addition, the details of the peak detection process will be described later.

[0050] Next, in S40, the processing device 10 extracts the peak frequency components detected in S30 from the FFT complex data S (t0, fb) of the K channel during the uplink tilt and the downlink tilt, respectively. Furthermore, the processing device 10 performs an arrival direction estimation process using an algorithm such as Multiple Signal Classification (hereinafter, MUSIC) on the extracted K peak frequency components during the uplink tilt, and estimates the direction of the target relative to the vehicle 80 for each target. Similarly, the processing device 10 performs an arrival direction estimation process on the extracted K peak frequency components during the downlink tilt, and estimates the direction of the target relative to the vehicle 80 for each target.

[0051] Next, in S50, the processing device 10 uses the azimuth of each target during the upward tilt and the downward tilt estimated in S40 and the power information of the synthetic complex data Sk (t0, fb) during the upward tilt and the downward tilt to match the peak frequency during the upward tilt and the peak frequency during the downward tilt corresponding to the same target in pairs. Furthermore, the processing device 10 calculates the speed v of the target relative to the vehicle 80 and the distance r from the vehicle 80 to the target based on the peak frequencies during the upward tilt and the downward tilt matched in pairs for each target.

[0052] Next, in S60, the processing device 10 tracks the target detected in this processing cycle. That is, the processing device 10 connects the information of the target detected until the previous processing cycle and the information of the target detected in this processing cycle, and calculates the moving direction of the target and the like.

[0053] Next, in S70, the processing device 10 sends target information including the target's speed v, distance r, direction, and moving direction to the ECU. The ECU uses the target information to execute an application such as driving assistance. For example, when the possibility of the vehicle 80 colliding with the target is relatively high, the ECU uses the target information to execute an application that outputs an alarm. This ends the present process.

[0054] In addition, in the present embodiment, the process of S20 corresponds to the function of the data calculation unit realized by the processing device 10 .

[0055] <2-2. Internal reflection>

[0056] Next, the reflection signal received by the radar device 100 will be described. Figure 6 As shown in FIG. 1 , the radar device 100 detects other vehicles 90 and pedestrians 95 that exist around the vehicle 80. However, in addition to the reflected waves from the targets, the radar device 100 also receives the internal reflected waves of the vehicle 80 that is not intended to be detected. The internal reflected waves are reflected waves generated inside the vehicle 80. Figure 6 As shown, the internal reflected waves include the reflected waves generated by the bumper 81. When the radar device 100 is mounted outside the bumper 81, the internal reflected waves include the reflected waves generated by the radome that protects the radar device 100 instead of the bumper 81. In addition, the internal reflected waves also include the reflected waves generated by the internal parts 82 of the vehicle 80. The transmission waves may be transmitted to the internal parts 82 via the bumper 81 and reflected. In addition, the reflected waves generated by the internal parts 82 may be transmitted to the radar device 100 via the bumper 81.

[0057] The relative position of the radar device 100 (specifically, the transmitting antenna 30 and the receiving antenna 40) and the bumper 81 and the internal parts 82 hardly changes in time. Specifically, the relative position changes only less than the order of mm in time. Therefore, the phase of the amplitude of the synthetic complex data Sk (t0, fb) representing the vehicle interior reflection signal hardly changes in time. On the other hand, the relative position of the radar device 100 and the target changes more than the order of mm in time. Therefore, the phase of the amplitude of the synthetic complex data Sk (t0, fb) representing the target reflection signal changes relatively greatly in time.

[0058] like Figure 7 As shown, in the power spectrum Ps of the synthetic complex data Sk(t0, fb), the peak pk_i indicating the vehicle interior reflection signal appears at a distance closer to the vehicle 80. Moreover, in the power spectrum Ps, at a position close to the peak pk_i, a peak pk_t indicating a close target existing at a position close to the radar device 100 also appears. Therefore, when the power value of the power spectrum Ps is compared with the detection threshold calculated based on the noise power and the peak is detected, both the peak pk_i and the peak pk_t are detected. It is difficult to distinguish whether the detected peak pk_i or pk_t is the peak indicating the vehicle interior reflection or the target.

[0059] In addition, in the present embodiment, the detection threshold is a detection threshold obtained based on noise power. Noise power is power other than reflection power from a target, such as thermal noise power of the radar device 100 and road surface reflection power. For example, when the target is set to only pedestrians, reflection power from other vehicles is also included in the noise power.

[0060] Here, as a method of suppressing static clutter components from a received signal, the following is proposed: Figure 8 The MTI processing shown in the figure. The MTI processing generates suppressed data Sm(t, fb) that suppresses signals without amplitude and phase changes by applying a high-pass filter to the time series of FFT complex data S(t, fb) by channel and by distance interval (i.e., by frequency band). For easy understanding, Figure 8 The high-pass filter process is replaced by a process of removing the average value of the time series during the period of t0-nT (n=0, 1, ...N-1) (i.e., the period of t0 to t0-(N-1)T) from the FFT complex data S (t0, fb) at time t0. N is a natural number representing the number of processing cycles included in the time series to which the high-pass filter process is applied.

[0061] Furthermore, after synthesizing the generated suppression data Sm(t0, fb) of each channel, the MTI process detects the detection threshold of the synthesized suppression data Smk(t0, fb). The detection threshold is detected, for example, by Constant False Alarm Rate (hereinafter, CFAR: Constant False Alarm Rate) detection. CFAR is a method of setting the detection threshold based on the signal power around the distance interval for setting the detection threshold.

[0062] Although MTI processing can reduce the signal to be suppressed when the amplitude or phase component of the signal to be suppressed slightly varies in time, it may not be able to suppress it completely. If it cannot be suppressed completely, the power of the signal to be suppressed may become larger than the detection threshold.

[0063] Here, the bumper 81, the interior parts 82, the transmitting antenna 30, and the receiving antenna 40 may have a small change of less than the order of mm due to the vibration accompanying the running of the vehicle 80. Therefore, the phase component of the amplitude of the complex data Sk (t, fb) representing the vehicle interior reflection signal may change slightly in time. Therefore, even if the MTI process is performed on the complex data Sk (t0, fb) at the time t0 calculated by the radar device 100, there is a possibility that the power of the internal reflection signal is larger than the detection threshold.

[0064] Therefore, in the present embodiment, the processing device 10 averages the time series of the FFT complex data S(t, fb) from time t0 to time t0-(N-1)T by channel and by frequency band to calculate the average data Sav(t0, fb). Furthermore, the processing device 10 synthesizes the average data Sav(t0, fb) of each channel to calculate the synthesized average data Savk(t0, fb), and feeds back the calculated synthesized average data Savk(t0, fb) to the detection threshold. That is, by increasing the detection threshold of the frequency band in which the vehicle interior reflection signal is observed according to the power of the vehicle interior reflection signal, the false detection of the vehicle interior reflection signal is suppressed. In the present embodiment, the synthesized average data Savk(t0, fb) is equivalent to the average data.

[0065] <2-3. Peak detection processing>

[0066] Next, refer to Fig. 9 The flowchart of FIG. 1 illustrates the procedure of the peak detection processing executed by the processing device 10.

[0067] First, in S300, the processing device 10 determines whether the vehicle 80 has moved during the averaging period described later based on the detection signal received from the vehicle speed sensor 70. If the vehicle 80 has not moved at all during the averaging period and has stopped, the process proceeds to S320, and if the vehicle 80 has moved during the averaging period, the process proceeds to S310.

[0068] In S310, the processing device 10 averages the FFT complex data S (t, fb) of a predetermined averaging period by channel and by frequency band to calculate the average data Sav (t0, fb). If the averaging period is set to a period of t0-nT (n=0, 1, ...N-1), the average data Sav (t0, fb) of each channel is expressed by the following formula (3). Furthermore, the processing device 10 synthesizes the average data Sav (t0, fb) of K channels to calculate the synthesized average data Savk (t0, fb). Specifically, the processing device 10 retains all the FFT complex data (t, fb) calculated in the past N processing cycles, and calculates the average of the N FFT complex data (t, fb).

[0069] Alternatively, the processing device 10 may calculate average data Sav(t0, fb) in the current processing cycle based on the latest FFT complex data (t0, fb) at time t0 and average data Sav(t0-1, fb) calculated in the previous processing cycle.

[0070] [Formula 3]

[0071]

[0072] like Fig.10 As shown in FIG. 1 , when the relative position between the radar device 100 and the reflector changes (specifically, changes of mm or more), the amplitude phase of the FFT complex data S (t, fb) changes randomly, so the size of the average vector becomes very small. As a result, the power of the average data Sav (t0, fb) is reduced by 10×log10 (N) compared to the power of the FFT complex data S (t0, fb).

[0073] On the other hand, Fig.11 As shown, when the relative position between the radar device 100 and the reflector is almost unchanged (specifically, the change is less than the mm level), the change in the phase of the amplitude of the FFT complex data S (t, fb) is small. Therefore, even if averaged, the magnitude and direction of the vector do not change much. As a result, the power of the average data Sav (t0, fb) is almost the same as the power of the FFT complex data S (t0, fb). Therefore, the power of the synthesized average data Savk (t0, fb) does not change in the frequency band of the observed vehicle internal reflection signal, and is greatly reduced in the frequency band of the observed target reflection signal, compared with the power of the synthesized complex data Sk (t0, fb).

[0074] In addition, when the vehicle 80 stops, since the relative position between the radar device 100 and the stationary object does not change, the phase of the amplitude of the FFT complex data S (t, fb) representing the stationary object does not change in time. Therefore, the processing device 10 includes the FFT complex data S (t, fb) calculated based on the discrete beat signal b (t) obtained when the vehicle 80 stops. If a plurality of FFT complex data S (t, fb) are averaged, the power of the synthetic average data Savk (t0, fb) representing the reflected signal from the stationary object is not suppressed. As a result, the detection threshold of the synthetic average data Savk (t0, fb) described later that is fed back rises in the frequency band of the observed stationary object, and the stationary object cannot be detected. Therefore, the processing device 10 calculates the average data Sav (t0, fb) and the synthetic average data Savk (t0, fb) only when the vehicle 80 moves during the averaging period.

[0075] Furthermore, the processing device 10 limits the distance range, that is, the range of the frequency band, for calculating the average data Sav(t0, fb) and the synthesized average data Savk(t0, fb) to a range based on the length of the vehicle 80. The internal reflection signal is observed within the range of the vehicle length from the front end to the rear end of the vehicle 80, and is not observed farther than the vehicle length. Therefore, the processing device 10 limits the distance range for calculating the average data Sav(t0, fb) and the synthesized average data Savk(t0, fb) to, for example, a range within the vehicle length of the vehicle 80, or a range within a value obtained by adding a margin value to the vehicle length of the vehicle 80.

[0076] In addition, the averaging period is the time during which the phase of the amplitude of the FFT complex data S (t, fb) of the target to be detected is considered to be random, for example, when the carrier frequency fc is 24 GHz, it can also be more than 1 second. In addition, if the averaging period is too long, when the power of the detected target reflection signal is high power, the power of the average data Sav (t, fb) in the frequency band will temporarily remain at a large value. Therefore, the averaging period is a period that does not extend the influence when a high-power target reflection signal is detected, for example, when the carrier frequency fc is 24 GHz, it can also be less than 10 seconds.

[0077] Next, in S320, the processing device 10 determines the detection threshold. Specifically, the processing device 10 adds the addition value X (dB) to the power value of the synthetic average data Savk (t0, fb) to calculate the first threshold as a frequency function. The addition value X (dB) is a predetermined positive value, which is a margin value set according to the accuracy of the synthetic average data Savk (t0, fb). In S310, when the synthetic average data Savk (t0, fb) is not calculated, the power value of the synthetic average data Savk (t0, fb) is 0.

[0078] In addition, the processing device 10 calculates the second threshold value based on the noise power during observation by the radar device 100. The processing device 10 adds an additional value Y (dB) to the total value of the noise power of each channel to calculate the second threshold value as a frequency function. The additional value Y (dB) is a predetermined value and is a margin value set according to the false detection rate.

[0079] Furthermore, the processing device 10 compares the first threshold and the second threshold according to the frequency band, and calculates the larger value as the detection threshold. Fig.12 As indicated by the double-dashed line, the calculated detection threshold is a first threshold near the radar device 100 observing the vehicle interior reflection signal, and is a value greater than the power of the synthetic complex data Sk (t0, fb) representing the vehicle interior reflection signal. In addition, the calculated detection threshold is a second threshold far from the radar device 100.

[0080] Next, in S330, the processing device 10 compares the synthesized complex data Sk(t, fb) with the detection threshold value determined in S320 for each frequency band, and detects the frequency at which the synthesized complex data Sk(t, fb) is greater than the detection threshold value and becomes maximum as a peak. This ends the present processing.

[0081] In addition, in this embodiment, the processing of S310 corresponds to the function of the average calculation unit implemented by the processing device 10, and the processing of S320 corresponds to the functions of the first threshold calculation unit, the second threshold calculation unit, and the detection threshold calculation unit implemented by the processing device 10. In addition, the processing of S330 corresponds to the function of the peak detection unit implemented by the processing device 10.

[0082] <3. Effect>

[0083] According to the first embodiment described above, the following effects can be obtained.

[0084] (1) A first threshold value of the synthetic average data Savk (t0, fb) fed back and a second threshold value calculated based on the noise power are compared, and the larger value is calculated as the detection threshold value. Thus, in the vicinity of the radar device 100 that observes the vehicle interior reflection signal, the detection threshold value is the first threshold value that is larger than the power value of the vehicle interior reflection signal. Therefore, it is possible to suppress erroneous detection of the vehicle interior reflection signal.

[0085] (2) Average data Sav(t0, fb) is calculated by averaging only the FFT complex data S(t0, fb) calculated from the discrete beat signal b(t) acquired when the vehicle 80 moves. This can suppress the inability to detect stationary objects around the vehicle 80.

[0086] (3) By limiting the distance range for calculating the average data Sav(t0, fb) and the synthesized average data Savk(t0, fb) to a range based on the vehicle length, it is possible to suppress erroneous detection of internal reflection signals without raising the threshold in the distance region where internal reflection does not occur in principle.

[0087] (Second Embodiment)

[0088] <1. Differences from the first embodiment>

[0089] Since the basic structure of the second embodiment is the same as that of the first embodiment, the description of the common structure is omitted and the description will be mainly focused on the differences. In addition, the same reference numerals as those of the first embodiment represent the same structure, and the previous description is referred to.

[0090] In the first embodiment described above, in the peak detection process, the synthesized average data Savk (t0, fb) is fed back to the detection threshold. In contrast, in the second embodiment, in the peak detection process, the synthesized average data Savk (t0, fb) is fed back to the detection threshold, and the synthesized complex data Sk (t0, fb) is updated to a value obtained by subtracting the synthesized average data Savk (t0, fb) from the synthesized complex data Sk (t0, fb) for each frequency band.

[0091] <2. Peak detection processing>

[0092] Next, refer to Fig.13 The peak processing performed by the processing device 10 of the second embodiment instead of the peak detection processing of the first embodiment is described with reference to the flowchart of FIG.

[0093] First, in S305 and S315, the processing device 10 performs Fig. 9 The processing of S300 and S310 is the same as that of S300.

[0094] Next, in S325, as shown in equation (4), the processing device 10 updates the FFT complex data S(t0, fb) to a subtraction value by channel and by frequency band. The subtraction value is a value obtained by subtracting the average data Sav(t0, fb) from the FFT complex data S(t0, fb).

[0095] S(t0, fb)=S(t0, fb)-Sav(t0, fb)…(4)

[0096] Furthermore, the processing device 10 synthesizes the FFT complex data S(t0, fb) of each channel and calculates the synthesized complex data Sk(t0, fb). As described above, the power of the average data Sav(t0, fb) is almost unchanged in the frequency band of the observation vehicle internal reflection signal, and is relatively greatly reduced in the frequency band of the observation target reflection signal, compared with the power of the FFT complex data S(t0, fb).

[0097] Fig.14 The left side of represents the threshold value obtained based on the FFT complex data S(t0, fb) before updating and the noise power. Fig.14 The right side of represents the threshold and detection threshold calculated based on the updated FFT complex data S(t0, fb) and noise power. Fig.14 As shown, the power of the updated FFT complex data S (t0, fb) is relatively greatly reduced in the frequency band of the observed vehicle interior reflection signal, but is hardly reduced in the frequency band of the observed target reflection signal, compared to before the update.

[0098] Next, in S335, the processing device 10 adds the addition value X (dB) to the power value of the synthetic average data Savk (t0, fb) to calculate the first threshold value as a frequency function. In this embodiment, the power of the updated FFT complex data S (t0, fb) is significantly reduced in the frequency band of the observed vehicle interior reflection signal compared to before the update, so the addition value X (dB) is a negative value.

[0099] In addition, the processing device 10 and Fig. 9 The second threshold is calculated in the same manner as in S320, and the first threshold and the second threshold are compared according to the frequency band, and the larger value is calculated as the detection threshold. Fig.14 As indicated by the double dotted line, the calculated detection threshold is a value that is greater than the power of the synthetic complex data Sk(t0, fb) representing the vehicle interior reflection signal near the radar device 100 and is smaller than the power of the synthetic complex data Sk(t0, fb) representing the target reflection signal.

[0100] Next, in S345, the processing device 10 and Fig. 9 In the same manner as the process of S330, the synthesized complex data Sk(t0, fb) calculated in S325 is compared with the detection threshold value determined in S335 for each frequency band to detect a peak. This process ends.

[0101] In addition, in the present embodiment, the process of S325 corresponds to the function of the update unit implemented by the processing device 10 .

[0102] <3. Effect>

[0103] According to the second embodiment described above, in addition to the effects (1) to (3) of the first embodiment described above, the following effects are obtained.

[0104] (4) The FFT complex data S(t0, fb) is updated to a value obtained by subtracting the average data Sav(t0, fb) from the FFT complex data S(t0, fb) for each channel and each frequency band. As a result, the FFT complex data S(t0, fb) indicating the internal reflection signal is suppressed, and the difference between the power value of the FFT complex data S(t0, fb) indicating the internal reflection signal and the power value of the FFT complex data S(t0, fb) indicating the target reflection signal increases. Furthermore, by adding a negative addition value X to the power value of the synthetic average data Savk(t0, fb), a first threshold value, which is a value between the power value of the FFT complex data S(t0, fb) indicating the internal reflection signal and the power value of the FFT complex data S(t0, fb) indicating the target reflection signal, is calculated. Therefore, it is possible to suppress erroneous detection of the vehicle interior reflection signal and appropriately detect the target reflection signal from the target located near the radar device 100.

[0105] (Other Embodiments)

[0106] As mentioned above, although the form for implementing this disclosure was described, this disclosure is not limited to the above-mentioned embodiment, and can be implemented with various modifications.

[0107] (a) In the above embodiment, the modulation method of the radar device 100 is the FMCW method, but the modulation method is not limited to the FMCW method. The present disclosure can be applied to radar devices using any modulation method (for example, multi-frequency CW method, pulse method) that extracts phase information.

[0108] (b) If a set of an upward ramp wave and a downward ramp wave is set as one transmission wave, the intervals between transmission waves may not be uniform. In other words, the intervals between processing cycles may not be uniform.

[0109] (d) In addition to the above-mentioned radar device, the present disclosure can also be implemented in various ways, such as a system with the radar device as a component, a program for causing a computer to function as a processing device of the radar device, a non-transitional physical recording medium such as a semiconductor memory recording the program, a peak detection method, etc.

Claims

1. A radar device mounted on a vehicle, The above radar device comprises: a transmitting antenna configured to transmit radar waves; a receiving antenna configured to receive a reflected wave generated by the reflection of the radar wave; a signal acquisition unit configured to acquire a received signal based on the reflected wave; a data calculation unit configured to perform frequency analysis on the received signal acquired by the signal acquisition unit and calculate complex data as a frequency function; an average calculation unit configured to calculate average data, wherein the average data is data obtained by averaging a plurality of the complex data in a predetermined period before a current processing cycle by frequency band; A first threshold value calculation unit is configured to add a predetermined additional value to the power value of the average data calculated by the average calculation unit to calculate a first threshold value as a frequency function; A second threshold value calculation unit is configured to calculate a second threshold value as a frequency function based on noise power in observation performed by the radar device; a detection threshold calculation unit configured to compare the first threshold calculated by the first threshold calculation unit and the second threshold calculated by the second threshold calculation unit according to the frequency band, and calculate the larger value as the detection threshold; as well as The peak detection unit is configured to compare the power value of the complex data in the current processing cycle with the detection threshold value according to the frequency band, and detect the frequency at which the power value of the complex data is greater than the detection threshold value and becomes maximum as a peak value, The average calculation unit is configured to calculate the average data on the condition that the vehicle moves during the predetermined period.

2. The radar device according to claim 1, wherein: The above addition value is negative. The radar device includes an updating unit configured to update the complex data calculated by the data calculating unit in the current processing cycle to a subtraction value, wherein the subtraction value is a value obtained by subtracting the average data calculated by the average calculating unit from the complex data in each frequency band. The peak detection unit is configured to compare the power value of the complex data updated by the update unit with the detection threshold value, and detect the frequency as the peak value.

3. The radar device according to claim 1 or 2, wherein: The frequency of the complex data calculated by the data calculation unit represents distance information from the radar device. The average calculation unit is configured to limit a distance range for calculating the average data to a range based on a length of the vehicle.

Images