Radar device

Abstract

The radar device of one aspect of the present disclosure has a transmission antenna (30), a reception antenna (40), a signal acquisition section (10, S10), an interference removal section (10, S30), a spectrum acquisition section (10, S40), a target extraction section (10, S80), and a first threshold value calculation section (10, S100). The interference removal section removes an interference signal from an acquired amplitude signal. The interference signal has an amplitude exceeding a first threshold value. The target extraction section extracts at least one target component exceeding a second threshold value from the spectrum. The first threshold value calculation section calculates the first threshold value using the extracted at least one target component.

Description

Technical Field

[0001] This disclosure relates to radar devices. Background Technology

[0002] There are instances where interference signals are mixed into the received signals received by radar devices. If the amplitude digital signal with overlapping interference signals is processed by FFT (Fast Fourier Transform), the noise floor of the FFT spectrum will rise, and the target signal will be buried by the noise floor and cannot be detected.

[0003] Non-Patent Document 1 below compares the amplitude value of the amplitude digital signal before FFT processing with a threshold, and removes the amplitude digital signal with an amplitude value greater than the threshold as an interference signal. This suppresses the rise of the noise floor in the FFT spectrum.

[0004] Non-Patent Literature 1: A Novel Iterative Inter-Radar Interference Reduction Scheme for Densely Deployed Automotive FMCW Radars (IRS2018) When the above threshold is set to a value smaller than the amplitude value of the target amplitude digital signal, the target signal is distorted, generating higher harmonics in the FFT spectrum. Summary of the Invention

[0005] One aspect of this disclosure is to provide a radar device capable of setting an appropriate threshold to remove interference signals from amplitude digital signals.

[0006] One aspect of the radar apparatus disclosed herein includes a transmitting antenna, a receiving antenna, a signal acquisition unit, an interference removal unit, a spectrum acquisition unit, a target extraction unit, and a first threshold calculation unit. The transmitting antenna is configured to transmit a transmitted wave in each processing cycle. The receiving antenna is configured to receive a reflected wave generated by the reflection of the transmitted wave. The signal acquisition unit is configured to acquire an amplitude signal corresponding to the reflected wave received by the receiving antenna. The interference removal unit is configured to remove interference signals from the amplitude signal acquired by the signal acquisition unit to update the amplitude signal. The interference signal has an amplitude exceeding the first threshold. The spectrum acquisition unit is configured to perform a Fourier transform on the amplitude signal to acquire a spectrum. The target extraction unit is configured to extract at least one target component exceeding a set second threshold from the spectrum acquired by the spectrum acquisition unit. The first threshold calculation unit is configured to calculate a first threshold using the at least one target component extracted by the target extraction unit.

[0007] According to one aspect of the radar apparatus of this disclosure, at least one target component exceeding a second threshold is extracted from the spectrum. In the amplitude signal, interference signals are mixed in during local time periods. Therefore, appropriately setting a first threshold and extracting the target signal from the amplitude signal is relatively difficult. On the other hand, in the spectrum, the power of the interference signal is dispersed across the entire frequency range, while the power of the target signal is concentrated at a certain inherent beat frequency. Therefore, appropriately setting a second threshold and extracting the target signal from the spectrum is relatively easy. Thus, in the aforementioned radar apparatus, the target component is extracted from the spectrum. The amplitude of the target signal contained in the amplitude signal can be inferred based on the extracted target component. That is, an appropriate first threshold corresponding to the amplitude of the target signal can be calculated based on the extracted target component. Therefore, using an appropriately set first threshold, interference signals can be removed from the amplitude signal. Attached Figure Description

[0008] Figure 1 This is a diagram showing the structure of the antenna device according to the first embodiment.

[0009] Figure 2 This is a diagram illustrating the situation that causes interference.

[0010] Figure 3 It is a graph showing the frequency changes of the transmitted wave, reflected wave, and interference wave relative to time.

[0011] Figure 4 This is a graph showing the changes of the target AD signal and the interfering AD signal over time.

[0012] Figure 5 It is a graph representing the target component and the interference component in the FFT spectrum.

[0013] Figure 6 This is a diagram representing the target AD signal and a properly set first threshold for removing interference.

[0014] Figure 7 This is a graph representing the target component in the FFT spectrum after interference removal using an appropriately set first threshold.

[0015] Figure 8 This is a diagram representing the target AD signal and an inappropriately set first threshold for removing interference.

[0016] Figure 9 This is a graph representing the target component in the FFT spectrum after interference removal using an inappropriately set first threshold.

[0017] Figure 10 This is a flowchart illustrating the signal processing of the first embodiment.

[0018] Figure 11 This is a diagram illustrating the amplitude estimation method of the target signal in the first embodiment.

[0019] Figure 12 This is a diagram illustrating the maximum amplitude of the target signal in the first embodiment.

[0020] Figure 13 This is a diagram illustrating the amplitude estimation method for the target signal in the second embodiment.

[0021] Figure 14 This is a diagram illustrating the amplitude estimation method for the target signal in the fourth embodiment.

[0022] Figure 15 This is a graph representing the target AD signal calculated using the target signal amplitude estimation method of the fourth embodiment. Detailed Implementation

[0023] The following description, with reference to the accompanying drawings, illustrates the methods used to implement this disclosure.

[0024] (First Implementation)

[0025] <1-1. Structure>

[0026] First, refer to Figure 1 The structure of the radar device 100 according to the first embodiment will be described. The radar device 100 uses Frequency Modulated Continuous Wave (FMCW) mode. Figure 2 As shown, the radar device 100 is mounted on the front bumper of the vehicle 70 at the center in the vehicle width direction.

[0027] The radar device 100 includes a processing unit 10, a tilt wave generator 20, a transmitting antenna 30, a receiving antenna 40, K mixers 50, and an analog-to-digital converter (ADC) 60.

[0028] The processing device 10 is centered around a microcomputer equipped with a CPU 11, ROM 12, RAM 13, and I / O. The processing device 10 generates a frequency control signal and sends it to the tilt wave generator 20. The frequency control signal sets the frequency of the transmitted signal. The tilt wave generator 20 generates a radar signal based on the frequency control signal received from the processing device 10 and sends the generated radar signal to the transmitting antenna 30. Furthermore, the tilt wave generator 20 supplies the generated radar signal to each of the K mixers 50.

[0029] The transmitting antenna 30 radiates a chirped wave modulated based on the radar signal received from the tilt wave generator 20. Specifically, as... Figure 3 As shown by the solid line, the transmitting antenna 30 repeatedly transmits a transmitted wave whose frequency monotonically changes from fc to fc+F during the modulation time Tc in each processing cycle.

[0030] The receiving antenna 40 includes K receiving antenna elements 41 (K is an integer greater than or equal to 2) arranged in a predetermined direction. Each receiving antenna element 41 receives the reflected wave generated by the radar wave being reflected by the target and supplies the reflected signal to the corresponding mixer 50.

[0031] K mixers 50 are each configured with one mixer for each receiving antenna element 41. Each mixer 50 mixes the radar signal supplied from the tilt wave generator 20 and the reflected signal supplied from the receiving antenna element 41 to generate a beat signal. The beat signal has the frequency difference between the radar signal and the reflected signal as its frequency component. The frequency difference between the radar signal and the reflected signal corresponds to the beat frequency. Each mixer 50 sends the generated beat signal to the ADC 60.

[0032] The ADC60 samples the beat signals of the K channels sent from the K mixers 50, generates amplitude digital signals (hereinafter, AD signals), and sends the generated AD signals of the K channels to the processing device 10. Specifically, the ADC60 is clock-synchronized with the tilt wave generator 20, and in each processing cycle, it begins sampling the beat signals after a predetermined time offset from the start of the transmitted wave, and samples over a certain time interval.

[0033] The processing unit 10 performs signal processing such as frequency analysis on the AD signals of K channels acquired from the ADC 60. The processing unit 10 loads and executes program code stored in a non-transitional physical recording medium such as ROM 12 via CPU 11 to implement the functions of a signal acquisition unit, an interference removal unit, a target extraction unit, a first threshold calculation unit, a rise determination unit, a peak extraction unit, and a transformation unit. The method of implementing these functions is not limited to software; some or all of the functions can also be implemented using hardware combining logic circuits, analog circuits, etc.

[0034] <1-2. Removal of Interference Signals>

[0035] Next, refer to Figures 2-9 The removal of interference signals is explained. For example... Figure 2As shown, interference can occur when other vehicles 80 are present near vehicle 70. Like vehicle 70, other vehicles 80 have a radar device 200 mounted in the center of their front bumper. The radar device 200 can be the same type as radar device 100, or it can be a different radar device. Pedestrians 90 are the detection targets of radar device 100.

[0036] Radar device 100 transmits a transmitted wave and receives a reflected wave from pedestrian 90. The transmitted wave is reflected by pedestrian 90, generating a reflected wave. Additionally, radar device 100 also receives a transmitted wave transmitted from radar device 200. The transmitted wave transmitted from radar device 200 acts as an interference wave relative to the reflected wave from the target.

[0037] like Figure 3 As shown, there are many cases where the slope of the interference wave from radar device 200 is not the same as the slope of the transmitted wave from radar device 100, and the period during which the frequencies of the interference wave and the transmitted wave intersect is limited. Therefore, as Figure 4 As shown, the period during which the interfering wave overlaps with the reflected wave is a relatively narrow period during which the frequencies of the interfering wave and the transmitted wave intersect. The reflected wave from the target attenuates according to the distance of the round trip between radar device 100 and the target. On the other hand, the interfering wave attenuates according to the distance of the one-way trip between radar device 200 and radar device 100. Therefore, as Figure 4 As shown, the amplitude of the AD signal based on the interference wave (hereinafter referred to as the interference AD ​​signal) is much larger than the amplitude of the AD signal based on the reflected wave from the target (hereinafter referred to as the target AD signal).

[0038] If a high-speed Fourier transform (FFT) is applied to an AD signal containing both interfering and target AD signals, an FFT signal containing the interfering component based on the interfering AD signal and the target component based on the target AD signal is calculated. For example... Figure 4 As shown, the power of the interference component is dispersed across the entire frequency range. On the other hand, the power of the target component is concentrated at the beat frequency.

[0039] like Figure 5 As shown, if the power of the interfering signal is large, the noise power increases across the entire frequency range, just as if the noise floor had increased in the FFT spectrum. As a result, the target component may be buried by noise and become undetectable.

[0040] Therefore, a method is proposed to remove interfering AD signals from the AD signal before FFT processing and suppress the rise of the noise floor in the FFT spectrum. Specifically, in the above method, the amplitude of the AD signal is compared with a first threshold, and components of the AD signal with amplitude values ​​exceeding the first threshold are identified as interfering AD signals and removed from the AD signal.

[0041] like Figure 6 As shown, when the first threshold is set to a value larger than the amplitude of the target AD signal, such as... Figure 7 As shown, the target component is calculated appropriately. However, if the first threshold is set to an excessively large value, interfering AD signals cannot be removed from the AD signal.

[0042] On the other hand, such as Figure 8 As shown, when the first threshold is set to a value smaller than the amplitude of the target AD signal, a portion of the target AD signal is also removed along with the interfering AD signal, resulting in distortion of the target AD signal. The result is as follows: Figure 9 As shown, the target component contains higher harmonic components.

[0043] Therefore, a first threshold can be appropriately set based on the amplitude value of the target AD signal. However, in order to infer the amplitude value of the target AD signal from the AD signal, it is necessary to store an amount of AD signal for one processing cycle in memory. Typically, an amount of AD signal for one processing cycle is not stored in memory, thus requiring an increase in memory capacity. On the other hand, the FFT signal after FFT processing is typically stored in memory. Therefore, if the FFT signal is used to infer the amplitude value of the target AD signal, it is not necessary to increase the memory capacity.

[0044] Therefore, in this embodiment, at least one target component is extracted from the FFT signal, and the extracted at least one target component is used to set a first threshold. That is, the amplitude value of the target AD signal corresponding to the extracted at least one target component is inferred, and the first threshold is set based on the inferred amplitude value of the target AD signal. Details regarding the setting of the first threshold will be described later.

[0045] <1-3. Signal Processing>

[0046] Next, refer to Figure 10 The flowchart below describes the signal processing performed by the processing device 10. The processing device 10 repeatedly performs this processing at a predetermined cycle.

[0047] First, in S10, the AD signals of K channels are obtained from ADC60.

[0048] Next, in S20, it is determined whether the interference determination flag was set to ON in the previous processing cycle. In this embodiment, as described later, the presence or absence of interference is determined based on the FFT signal. Therefore, the determination result from the previous processing cycle is used. It can be assumed that the interference situation does not change drastically during consecutive processing cycles. If it is determined in S20 that the interference determination flag is set to ON, the process proceeds to S30.

[0049] In S30, a first threshold set in the previous processing cycle is used to remove interfering AD signals from the AD signal. The first threshold is a positive value. Specifically, as shown in equation (1), the amplitude value of each sampled signal contained in the AD signal is compared with the magnitude of the first threshold, and sampled signals with amplitude values ​​exceeding the magnitude of the first threshold are removed from the AD signal. Here, m represents the sample number, n represents the chirp number of the multiple chirps transmitted in each processing cycle, ch represents the receiving channel number from 0 to K-1, ad(m,n,ch) represents the AD signal, Th1(ch) represents the first threshold, and ad_rej(m,n,ch) represents the updated AD signal after removing interference. The first threshold Th1(ch) is set for each receiving channel in the processing described later.

[0050] Furthermore, as shown in equation (1), in the AD signal, instead of the sampled signal with an amplitude value exceeding the first threshold Th1(ch), a sampled signal with an amplitude of the first threshold Th1(ch) is added to update the AD signal. That is, in the AD signal, the amplitude value is limited to the first threshold Th1(ch) during periods when the amplitude value is greater than the first threshold Th1(ch), and the amplitude value is limited to the first threshold Th1(ch) × (-1) during periods when the amplitude value is less than the first threshold Th1(ch). After that, the process proceeds to S40.

[0051] [Formula 1]

[0052]

[0053] Alternatively, as shown in Equation (2), the amplitude values ​​of each sampled signal included in the AD signal are compared with the magnitude of the first threshold Th1(ch), and sampled signals with amplitude values ​​exceeding the magnitude of the first threshold Th1(ch) are removed. Furthermore, as shown in Equation (2), in the AD signal, sampled signals with amplitude values ​​exceeding the magnitude of the first threshold Th1(ch) are added instead of sampled signals with amplitude values ​​exceeding the magnitude of the first threshold Th1(ch). That is, in the AD signal, periods with amplitudes greater than the first threshold Th1(ch) are filled with 0.

[0054] [Equation 2]

[0055]

[0056] Furthermore, while zero-filling offers greater interference suppression than limiting, it can increase higher harmonic components if the target AD signal is incorrectly removed. Therefore, either zero-filling or limiting can be appropriately chosen based on whether interference suppression or higher harmonic suppression is prioritized.

[0057] On the other hand, if it is determined in S20 that the interference determination flag is set to Off, the processing in S30 is skipped, and the process proceeds to S40. That is, by avoiding performing unnecessary interference removal processing in an interference-free environment, the possibility of distortion of the target AD signal is reduced.

[0058] In S40, the AD signal is processed by FFT for each received channel to calculate the FFT signal. If interference removal processing is performed in S30, the updated AD signal in S30 is processed by FFT. Alternatively, if interference removal processing is not performed in S30, the AD signal acquired in S10 is processed by FFT. Specifically, the AD signal is subjected to a two-dimensional Fourier transform using the following equation (3). If interference removal processing is not performed in S30, ad_rej(m,n,ch) in equation (3) is set to ad(m,n,ch). P represents the distance interval number from 0 to M-1, q represents the velocity interval number from 0 to N-1, fft(p,q,ch) represents the FFT signal, and wind(m,n) represents the window function. M and N are integers greater than or equal to 2.

[0059] [Formula 3]

[0060]

[0061] Next, in S50, interference determination is performed. Specifically, equations (4) to (7) below are used to determine whether there is an increase in noise power. First, using equation (4), the total power pow_ch_sum(p,q) is calculated by summing the power of the FFT signal fft(p,q,ch) of the total received channel quantity.

[0062] [Formula 4]

[0063] pow_ch_sum(p, q)=∑abs(fft(p, q, ch)) 2 …(4) Next, using equation (5), the median average power pow_median of the calculated total power pow_ch_sum(p,q) is calculated.

[0064] [Formula 5]

[0065] pow_median=10*log10(pow_ch_sum(p,q))…(5) Next, calculate the minimum power pow_min_hold of the median average power pow_median in the X1 processing cycles prior to the current processing cycle. X1 is, for example, 100. The minimum power pow_min_hold is equivalent to the thermal noise of the radar device 100, specifically, the thermal noise in the receiving path from the receiving antenna 40 to the processing device 10. Since interference may not be observed in every processing cycle, the median average power pow_median in the long-term processing cycles includes the median average power pow_median of non-overlapping interference. The median average power pow_median of non-overlapping interference is smaller than the median average power pow_median of overlapping interference. Therefore, by selecting the minimum power pow_min_hold from the median average power pow_median in the long-term processing cycles, the thermal noise of the radar device 100 can be calculated.

[0066] Next, it is determined which of the interference determination opening and closing conditions is met. Specifically, if equation (6) is satisfied Y times in the previous X2 processing cycles, it is determined that the interference determination opening condition is met. For example, X2 is 10, Y is 5, and pow_thre is 10dB.

[0067] That is, if, in a processing cycle of X2 times, the difference between the median average power pow_median and the minimum power pow_min_hold is greater than pow_thre for Y or more times, the interference detection enable condition is determined to be met. Furthermore, it is determined that there is an increase in noise power caused by interference. If the interference detection enable condition is met, the process proceeds to S60, where the interference detection flag is set to be enabled, and then proceeds to S80.

[0068] [Formula 6]

[0069] pow_median>pow_min_hold+pow+thre…(6)

[0070] Furthermore, if equation (7) is satisfied continuously in the previous X3 processing cycles, the interference determination shutdown condition is determined to be met. X3 is, for example, 5. That is, if the difference between the median average power pow_median and the minimum power pow_min_hold is less than pow_thre for more than X3 processing cycles, the interference determination shutdown condition is met. Furthermore, it is determined that there is no increase in noise power caused by interference. If the interference determination shutdown condition is determined to be met, the process proceeds to S70, the interference determination flag is set to off, and the process proceeds to S80.

[0071] [Formula 7]

[0072] pow_median<pow_min_hold+pow+thre(7) Next, in S80, for each received channel, at least one target component is extracted from the FFT spectrum calculated in S40, and a peak component is extracted from each extracted target component. Specifically, using the following equation (8), a group of distance interval numbers p and velocity interval numbers q that have a total power pow_ch_sum(p,q) greater than the second threshold Th2 is searched from the FFT spectrum as target components. That is, a group of intervals (p,q) that satisfy equation (8) is searched. For example, the second threshold Th2 is calculated by adding a predetermined value to the median average power pow_median. The predetermined value is, for example, 5dB.

[0073] [Formula 8]

[0074] pow_ch_sum(p, q)>Th2…(8) Since many interfering AD signals are observed in a short period of time, the interfering components corresponding to the interfering AD signals are widely distributed throughout the entire frequency range of the FFT spectrum. Therefore, in the FFT spectrum, there are almost no cases where the interfering components exceed the second threshold Th2. In addition, even if the interfering components exceed the second threshold, since only a part of the power of the interfering components exceeds the second threshold, it will not have a significant impact on the estimation of the amplitude of the target AD signal described later.

[0075] On the other hand, the target component corresponding to the target AD signal is only distributed near the beat frequency of the FFT spectrum. Therefore, the power of the target component exceeds the second threshold Th2. Therefore, the detection accuracy of the target component separated from the FFT signal is better than that of the target AD signal separated from the AD signal.

[0076] Next, using equation (9) below, the group of intervals (p, q) with the maximum power is searched from the group of intervals (p, q) that are the extracted target components as the peak components. That is, the group of intervals (p, q) that satisfy all the equations in equation (9) below is searched.

[0077] [Formula 9]

[0078]

[0079] Furthermore, peak extraction processing in S80 is part of the signal processing of a typical radar device and is not an additional process added to remove interference. Therefore, the processing time is not increased by performing peak extraction processing.

[0080] Next, in S90, the amplitude value of the target AD signal is inferred for each received channel. Specifically, the amplitude value of the AD signal corresponding to the additive signal is calculated as the amplitude prediction value amp_est(ch) using the following equation (10). The additive signal is the FFT signal from which the multiple peak components extracted in S80 are superimposed with the same phase. The initial value of the amplitude prediction value amp_est(ch) is set to 0, and amp_est(ch) is updated to reflect the number of peaks extracted in S80. Figure 11 As shown, for example, when three peak values ​​PK1, PK2, and PK3 are extracted, the amplitude prediction value amp_est(ch) is updated three times. p_peak and q_peak represent the velocity interval number and distance interval number that satisfy equations (8) and (9), respectively. Additionally, in Figure 11 And the following Figure 13 , Figure 14 In the diagram, Tg1, Tg2, and Tg3 represent the target components, respectively.

[0081] [Formula 10]

[0082] amp_est(ch)+=abs(fft(p_peak, q_peak, ch)) / (MN)(10) such as Figure 12 As shown, when the target AD signal is composed of a first AD signal S1 and a second AD signal S2, the amplitude of the target AD signal is maximized when the peaks of the first AD signal S1 and the second AD signal S2 overlap. That is, the amplitude of the AD signal corresponding to the additive signal after the multiple peak components contained in the target component overlap with the same phase is equivalent to the maximum possible amplitude of the target AD signal. In S90, the amplitude prediction value amp_est(ch) is calculated as the maximum possible amplitude of the target AD signal.

[0083] Next, in S100, as shown in Equation (11), for each receiving channel, the amplitude prediction value amp_est(ch) calculated in S90 is multiplied by the coefficient coef to calculate the first threshold Th(ch). The coefficient coef is a value greater than or equal to 1. The larger the value of the coefficient coef, the smaller the interference suppression effect, but the lower the possibility of erroneously removing the target signal. The value of the coefficient coef can be appropriately set to be greater than or equal to 1.

[0084] [Equation 11]

[0085] Th1(ch) = coef * amp_est(ch) (11) Next, in S110, the peak components extracted in S80 are used to perform processing to assist the driving of vehicle 70. Specifically, target orientation estimation, tracking, clustering, and other processing are performed. This concludes the processing.

[0086] <1-4. Effects>

[0087] According to the first embodiment described above, the following effects can be obtained.

[0088] (1) Extract at least one target component exceeding the second threshold Th2 from the FFT spectrum. Furthermore, infer the amplitude value of the target AD signal based on the extracted target component. Then, calculate the first threshold Th1(ch) based on the inferred amplitude value. Finally, use the calculated first threshold Th1(ch) to remove interfering AD signals from the AD signal. Therefore, it is possible to remove interfering AD signals from the AD signal using an appropriately set first threshold Th1(ch).

[0089] (2) When using the first threshold Th1(ch) calculated in the current processing cycle to remove interference signals, the AD signal needs to be stored in memory. Furthermore, in the current processing cycle, after calculating the first threshold Th1(ch), interference signals need to be removed from the AD signal stored in memory, and the FFT spectrum is recalculated based on the AD signal after interference removal. Therefore, the required memory capacity increases, and the processing time of each processing cycle increases. In contrast, by using the first threshold Th1(ch) calculated in the previous processing cycle, the required memory capacity can be reduced, and the processing time of each processing cycle can be reduced.

[0090] (3) In each processing cycle, it is determined whether there is an increase in noise power. Only when an increase in noise power is determined, the process of removing interfering AD signals from the AD signal is performed. This reduces the risk of distortion of the target AD signal.

[0091] (4) By performing a process to remove the interfering AD signal from the AD signal when the noise power is determined to be rising in the previous processing cycle, the processing time in each processing cycle can be suppressed.

[0092] (5) By performing amplitude limiting when removing interference signals, it is possible to suppress high-order harmonic components even if the target AD signal is mistakenly removed.

[0093] (6) By making multiple peak components overlap with the same phase, the maximum amplitude value that can become the amplitude value of the AD signal can be calculated as the amplitude prediction value amp_est(ch). Moreover, by calculating the first threshold Th1(ch) based on the calculated amplitude prediction value amp_est(ch), the erroneous removal of the target AD signal can be suppressed.

[0094] (Second Implementation)

[0095] <2-1. Differences from the first embodiment>

[0096] Since the basic structure of the second embodiment is the same as that of the first embodiment, the description of the common structures is omitted, and the description focuses on the differences. Furthermore, the same reference numerals as in the first embodiment denote the same structures, as described above.

[0097] In the first embodiment described above, in S90, the processing device 10 infers the amplitude value of the target AD signal based on multiple peak components. In contrast, in the second embodiment, in S90, the processing device 10 uses not only multiple peak components but also all FFT signals contained in the multiple target components to infer the amplitude value of the target AD signal, which differs from the first embodiment.

[0098] <2-2. Amplitude estimation of the target AD signal>

[0099] Next, the amplitude estimation processing of the target AD signal performed by the processing device 10 in the second embodiment will be described. In the second embodiment, using equation (12), the amplitude value of the AD signal corresponding to the sum of the FFT signals of the multiple intervals (p, q) that satisfy equation (8) with the same phase overlap is calculated as the amplitude estimation value amp_est(ch). The initial value of the amplitude estimation value amp_est(ch) is 0, and the number of the intervals (p, q) that satisfy equation (8) is updated by the amplitude estimation value amp_est(ch). p_above, q_above represent the velocity interval number and the distance interval number that satisfy equation (8).

[0100] like Figure 13As shown, when the target component contains 9 intervals Bin1 to Bin9, the amplitude prediction value amp_est(ch) is updated 9 times. Among the 9 intervals Bin1 to Bin9, there are 3 intervals Bin3, Bin7, and Bin9 that correspond to the peak value.

[0101] [Equation 12]

[0102] amp_est(ch)+=abs(fft(p_above, q_above, ch)) / (MN)(12)

[0103] <2-3. Effects>

[0104] According to the second embodiment described above, it achieves the same effect as the first embodiment described above (1) to (6).

[0105] (Third Implementation)

[0106] <3-1. Differences from the first embodiment>

[0107] Since the basic structure of the third embodiment is the same as that of the first embodiment, the description of the common structures is omitted, and the description focuses on the differences. Furthermore, the same reference numerals as in the first embodiment denote the same structures, as described above.

[0108] In the first embodiment described above, in S90, the processing device 10 calculates an amplitude prediction value amp_est(ch) based on an additive signal consisting of multiple peak components overlapping with the same phase, corresponding to the maximum possible amplitude value of the target AD signal. In contrast, in the third embodiment, in S90, the processing device 10 assumes that the target AD signal is a single frequency signal, and the point at which it calculates the amplitude prediction value amp_est(ch) differs from that in the first embodiment.

[0109] <3-2. Amplitude estimation of the target AD signal>

[0110] Next, the estimation processing of the amplitude value of the target AD signal performed by the processing device 10 in the third embodiment will be described. In the third embodiment, for each receiving channel, the power values ​​in the group (p, q) of the interval satisfying equation (8) are added together using equation (13) to calculate the additive power value pow_est(ch). The initial value of the additive power value pow_est(ch) is 0, and the number of groups (p, q) of the interval satisfying equation (8) is updated by the additive power value pow_est(ch).

[0111] [Equation 13]

[0112] pow_est(ch)+=abs(fft(p_above, q_above, ch)) 2 (13) Next, assuming that the target AD signal is a single-frequency signal, use Equation (14) to calculate the amplitude value of the single-frequency AD signal that is equivalent to the additive power value pow_est(ch) as the amplitude prediction value amp_est(ch).

[0113] [Formula 14]

[0114] amp_est(ch) +=sprt(2*pow_est(ch)) / (MN)(14) In the first embodiment, amp_est(ch) is used as the amplitude prediction value to predict the amplitude of the target AD signal when all peak components are in phase. In the second embodiment, amp_est(ch) is used as the amplitude prediction value to predict the amplitude of the target AD signal when all FFT signals included in the target component are in phase. Therefore, in both the first and second embodiments, the amplitude prediction value amp_est(ch) is often larger than the actual amplitude of the target AD signal.

[0115] On the other hand, in the third embodiment, it is assumed that the target AD signal is a single-frequency signal, and the amplitude value of the single-frequency AD signal equivalent to the additive power value pow_est(ch) is calculated as the amplitude prediction value amp_est(ch). The assumption that the target AD signal is a single-frequency signal holds true if one of the reflected signals in the multiple reflected waves observed by the radar device 100 is sufficiently larger than the others. Furthermore, generally speaking, the reflected signals received by the radar device 100 have the characteristic that one of the multiple reflected signals is larger than the others. Under the assumptions stated above, the prediction accuracy of the amplitude prediction value amp_est(ch) in the third embodiment is higher than that in the first and second embodiments.

[0116] <3-3. Effects>

[0117] According to the third embodiment described above, in addition to the effects (1) to (5) of the first embodiment described above, the following effects are also achieved.

[0118] (7) When one of the multiple reflected signals observed by the radar device 100 is significantly larger than the others, it is possible to assume that the target AD signal is a single-frequency signal and to accurately predict the amplitude of the target AD signal. Furthermore, it is possible to set a first threshold Th1(ch) with good accuracy.

[0119] (Fourth Implementation)

[0120] <4-1. Differences from the first embodiment>

[0121] Since the basic structure of the fourth embodiment is the same as that of the first embodiment, the description of the common structures is omitted, and the description focuses on the differences. Furthermore, the same reference numerals as in the first embodiment denote the same structures, as described above.

[0122] In the first embodiment described above, in S90, the amplitude prediction value amp_est(ch) is calculated based on the additive signal after multiple peak components are overlapped with the same phase. In contrast, in the fourth embodiment, the point at which the inverse Fourier transform is performed on all FFT signals contained in the target component to calculate the amplitude prediction value amp_est(ch) differs from that in the first embodiment.

[0123] <4-2. Amplitude estimation of the target AD signal>

[0124] Next, in the fourth embodiment, the estimation processing of the amplitude value of the target AD signal performed by the processing device 10 will be described. In the fourth embodiment, for each receiving channel, an extraction signal fft_above(p,q,ch) from which the target component has been extracted from the FFT spectrum is generated using equation (15). The extraction signal fft_above(p,q,ch) is equivalent to the following signal: in each receiving channel, the values ​​of the FFT signals in the group (p,q) where the total power value of the FFT signals exceeds the second threshold Th2 remain unchanged, and the values ​​of the FFT signals in the group (p,q) where the values ​​of the FFT signals are below the second threshold Th2 are 0. That is, as Figure 14 As shown by the dashed line, in the amplitude prediction processing of the fourth embodiment, FFT signals exceeding the second threshold Th2 are extracted from each target component.

[0125] [Formula 15]

[0126] fft_above(p, Q, ... c h)=fft(p,Q,ch)·…(15)

[0127] ==0

[0128] Next, using equation (16), the generated extracted signal fft_above(p,q,ch) is subjected to an inverse Fourier transform to calculate the inverse Fourier signal ad_above(m,n,ch). For example... Figure 15 As shown, the inverse Fourier signal ad_above(m,n,ch) is equivalent to the AD signal after removing the interference AD ​​signal.

[0129] [Formula 16]

[0130]

[0131] Furthermore, as shown in Equation (17), the maximum values ​​of m and n in the inverse Fourier signal ad_above(m,n,ch) are calculated as the amplitude prediction values ​​amp_est(ch) in each receiving channel.

[0132] [Equation 17]

[0133] amp_est(ch)=max(abs(ad_above(m,n,ch)))…(17)

[0134] The calculation of the amplitude estimation value amp_est(ch) in the fourth embodiment is time-consuming compared to the calculation of the amplitude estimation value amp_est(ch) in the first to third embodiments. However, in the fourth embodiment, it is not assumed that all FFT signals are in the same phase as in the first and second embodiments. Furthermore, in the fourth embodiment, it is not assumed that one of the multiple reflected signals is larger than the others, as in the third embodiment. Therefore, the estimation accuracy of the amplitude estimation value amp_est(ch) in the fourth embodiment is higher than that in the first to third embodiments.

[0135] <4-3. Effects>

[0136] According to the fourth embodiment described above, in addition to the effects (1) to (5) of the first embodiment described above, the following effects are achieved.

[0137] (8) The target AD signal, from which interfering AD signals have been removed, is calculated by extracting the FFT signals contained in multiple target components and performing inverse Fourier transform. The first threshold Th1(ch) can be calculated with high precision based on the maximum value in the calculated target AD signal.

[0138] (Other implementation methods)

[0139] The above describes the methods for implementing this disclosure, but this disclosure is not limited to the above-described embodiments and can be implemented in various ways.

[0140] (a) The radar device 100 and method described in this disclosure can also be implemented using a dedicated computer, which is provided by comprising a processor programmed to perform one or more functions embodied in a computer program and a memory. Alternatively, the radar device 100 and method described in this disclosure can also be implemented using a dedicated computer provided by employing one or more dedicated hardware logic circuits to construct a processor. Alternatively, the radar device 100 and method described in this disclosure can also be implemented using one or more dedicated computers, which are composed of a combination of a processor programmed to perform one or more functions and a memory, and a processor composed of one or more hardware logic circuits. In addition, the computer program can also be stored as instructions executed by the computer on a computer-readable non-transitional tangible recording medium. In the method of implementing the functions of the various parts included in the radar device 100, it is not necessary to include software, and one or more hardware components can be used to implement all its functions.

[0141] (b) Multiple functions of one component in the above embodiments can be achieved through multiple components, or one function of one component can be achieved through multiple components. Alternatively, multiple functions of multiple components can be achieved through one component, or one function achieved by multiple components can be achieved through one component. Furthermore, a portion of the structure of the above embodiments can be omitted. Additionally, at least a portion of the structure of other above embodiments can be added to or replaced.

[0142] (c) In addition to the radar device described above, this disclosure can also be implemented in various ways, such as a system that includes the radar device, a program that enables a computer to function as the radar device, a non-transitional physical recording medium such as a semiconductor memory that records the program, and an interference removal method.

Claims

1. A radar device comprising: The transmitting antenna is configured to transmit a wave in each processing cycle; The receiving antenna is configured to receive the reflected wave generated by the reflection of the transmitted wave. The signal acquisition unit is configured to acquire an amplitude signal corresponding to the reflected wave received by the receiving antenna. The interference removal unit is configured to remove interference signals from the amplitude signal acquired by the signal acquisition unit to update the amplitude signal, wherein the interference signals have an amplitude value exceeding a first threshold. The spectrum acquisition unit is configured to perform a Fourier transform on the above-mentioned amplitude signal to acquire the spectrum; The target extraction unit extracts at least one target component exceeding a set second threshold from the spectrum acquired by the spectrum acquisition unit; and The first threshold calculation unit uses at least one target component extracted by the target extraction unit to calculate the first threshold.

2. The radar device according to claim 1, wherein, The interference removal unit is configured to use the first threshold calculated by the first threshold calculation unit in the previous processing cycle.

3. The radar device according to claim 1 or 2, wherein, It also includes an increase determination unit, which is configured to determine whether the noise power in the spectrum acquired by the spectrum acquisition unit has increased. The interference removal unit is configured to update the amplitude signal when the rise determination unit determines that there is an increase in noise power.

4. The radar device according to claim 3, wherein, The interference removal unit is configured to update the amplitude signal if the rise determination unit determines that there is an increase in noise power in the previous processing cycle.

5. The radar device according to any one of claims 1 to 4, wherein, The interference removal unit is configured to replace the removed interference signal with a signal having an amplitude value of the first threshold value in the amplitude signal.

6. The radar device according to any one of claims 1 to 5, wherein, It also includes a peak extraction unit, which is configured to extract at least one peak component from the at least one target component extracted by the target extraction unit. The first threshold calculation unit is configured to calculate the first threshold based on an additive signal, which is equivalent to a signal in which at least one peak component extracted by the peak extraction unit overlaps with the same phase.

7. The radar device according to any one of claims 1 to 5, wherein, The first threshold calculation unit is configured to calculate the first threshold based on an additive signal, which is equivalent to a signal that causes at least one target component extracted by the target extraction unit to overlap with the same phase.

8. The radar device according to any one of claims 1 to 5, wherein, The first threshold calculation unit is configured to calculate the first threshold based on the total power of the at least one target component extracted by the target extraction unit.

9. The radar device according to any one of claims 1 to 5, wherein, It also includes a transformation unit configured to perform an inverse Fourier transform on at least one target component extracted by the target extraction unit to calculate an inverse Fourier signal. The first threshold calculation unit is configured to calculate the first threshold based on the maximum value in the inverse Fourier signal calculated by the transformation unit.

Images