A method for suppressing interference signals in fast chirp FMCW radar sensors for automotive applications.

Abstract

A method for suppressing an interfering signal having a known interfering signal frequency spectrum in an automotive fast chirp FMCW radar sensor, comprising: subjecting a baseband signal formed by mixing a transmitted signal with a received signal to a discrete Fourier transform in at least two dimensions, one of the at least two dimensions representing a Doppler spectrum, i.e., a spectrum of Doppler frequencies resulting from the radial velocity of a radar target relative to the radar sensor; and determining the start times (t, ..., t) of chirps (28) in the transmitted signal. n ) is the time interval ΔT where the distance of the chirp (28) varies nonlinearly. n In the method, the time intervals Δt n The fluctuation width Δ of at least one interference signal frequency f in the Doppler spectrum depends on the interference signal frequency spectrum. i is intentionally selected to be suppressed.

Description

[Technical Field]

[0001] The present invention provides a method for suppressing an interfering signal having a known interfering signal frequency spectrum in a fast chirp FMCW radar sensor for automotive applications, the method comprising: subjecting a baseband signal formed by mixing a transmitted signal with a received signal to a discrete Fourier transform in at least two dimensions, one of the at least two dimensions representing the Doppler spectrum, i.e., the spectrum of Doppler frequencies resulting from the radial velocity of the radar target relative to the radar sensor; and determining whether the chirp start time in the transmitted signal is within a time interval Δt during which the chirp range varies nonlinearly. n The present invention relates to a method in which only the first and second eigenvalues ​​are selected to differ from each other. [Background technology]

[0002] In FMCW (Frequency Modulated Continuous Wave) radar, the frequency of the transmitted radar signal is modulated in a ramp pattern. When a received radar echo mixes with a portion of the signal transmitted at the same time, the frequency difference between the two signals creates a beat, resulting in a baseband signal with a frequency corresponding to the frequency difference between the transmitted and received signals. This frequency difference is proportional to the signal propagation time and, therefore, to the distance of the located radar target, so the frequency of the baseband signal contains information about the object's distance. However, for a moving radar target, the frequency of the baseband signal also depends on the relative velocity between the radar target and the radar sensor due to the Doppler effect.

[0003] In a fast-chirp FMCW radar sensor, very steep frequency ramps, or chirps, are transmitted in rapid succession. A large ramp gradient enhances the sensor's range sensitivity. However, the ramp gradient does not affect the frequency shift due to the Doppler effect; therefore, if the ramp gradient is large enough, the Doppler effect can be ignored compared to the range-dependent frequency shift. In the Fourier spectrum, each located object is represented by a peak at a specific frequency, which indicates the object's range to a good approximation. To obtain information about relative velocity, the radar echoes acquired from different frequency ramps exhibit a phase shift that depends on the relative velocity. A second Fourier transform, in which the signals received at corresponding support points in successive frequency ramps, rather than those acquired during a single ramp, is transformed, yields a two-dimensional Fourier spectrum, where the first dimension indicates the object's range and the second dimension indicates the Doppler frequency and thus the relative velocity.

[0004] DE102009016480B4 describes a radar sensor that operates according to this principle, in which the time distance between successive chirps varies according to a random sequence. This variation in time distance, in combination with a special method for object tracking, improves the object's validity and the resolution of ambiguity when determining the relative velocity. However, it is also noted that as a side effect, the variation in time distance can make radar localization more robust against interfering signals. Summary of the Invention [Problem to be solved by the invention]

[0005] The object of the present invention is to improve the suppression of interfering signals in scenarios where these signals have a known frequency or frequency distribution and are scattered into a radar sensor. [Means for solving the problem]

[0006] This problem arises when using radar sensors of the type mentioned at the beginning, with a time interval Δtn The fluctuation width Δ of at least one interference signal frequency f in the Doppler spectrum depends on the interference signal frequency spectrum. i is intentionally selected to be suppressed, thereby solving the problem according to the present invention.

[0007] The present invention takes advantage of the fact that variations in the distance between chirps cause phase shifts in the baseband signal, which, upon Fourier transformation, translate into frequency shifts in the Doppler spectrum. Essentially, this effect broadens and flattens not only the peaks due to interfering signals in the Doppler spectrum, but also the peaks due to the useful signal, which indicates the relative velocity of the actual radar target. For a signal with a given frequency, however, the flattening of the signal peaks depends not only on the variation in the chirp distance, but also on the frequency of the signal of interest. Therefore, if the frequency of the interfering signal is known, the variation can be adjusted to intentionally suppress the signal peaks due to the interfering signal, while leaving the spectrum of the useful signal, which typically has a different frequency, largely unaffected, thereby improving the detectability of the actual radar target.

[0008] The present invention can be used, for example, to suppress interference signals resulting from a clock voltage regulator used to power a radar sensor and / or other electronic components, e.g., in a vehicle. The switching frequency of the voltage regulator generates ripple in the supply voltage for the radar electronics. In the highly sensitive circuit components of the radar sensor's transmit and receive circuits, this can result in coupling of the interference frequency into the transmit or receive signal, ultimately impairing the readability of the received signal. A known method for reducing the effects of such interference is to use a passive filter (LC filter) to attenuate the amplitude of the ripple to such an extent that the interference can no longer be detected in the spectrum of the received signal. However, due to the high sensitivity of the transmit and receive circuit components, the requirements for filtering are very high, and therefore this type of interference suppression is very complex.

[0009] A further known method for reducing the maximum interference power at the switching frequency and its harmonics is to vary the switching frequency over time in order to spread the power spectrally (spread spectrum method). However, the possibilities for varying the switching frequency of a voltage regulator are generally very limited due to boundary conditions that must be observed, and therefore this method does not always work. In contrast, the method according to the invention has the advantage that the interference frequency does not have to be spread, but instead only the relevant peak in the Doppler spectrum is spread.

[0010] However, the method according to the invention can also be used to suppress other interfering signals, whose frequencies are known and which typically occur around radar sensors in motor vehicles. Advantageous embodiments and further developments of the invention are defined in the dependent claims.

[0011] In some known high-speed chirp radar sensors, chirps are grouped into packets separated from each other by a time gap. In this case, the start time of each packet can be varied, and the distance between chirps within the packet can be uniform. The variation in the time distance is limited only by the minimum size of the time gap and can therefore be significantly greater than the time distance between two consecutive chirps within a packet. Naturally, the distance between chirps within a packet can also be varied, albeit with a smaller variation. The two different variation ranges can then be used, for example, to suppress two different interfering signals.

[0012] The subject of the invention is also a radar sensor in which the above-described method is implemented, the variation width of the time interval being adapted to the frequency of an internal interference signal source of the radar sensor. Furthermore, the subject of the invention is a motor vehicle equipped with a radar sensor in which the above-described method is implemented and in which the variation width of the time intervals is adapted to the frequency of an interfering signal source present in the motor vehicle.

[0013] Exemplary embodiments of the invention will now be explained in more detail on the basis of the drawings. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a block diagram of an FMCW radar sensor to which the present invention can be applied. [Figure 2] FIG. 1 is a collection of graphs illustrating how a fast chirp FMCW radar sensor functions. [Figure 3] 3 shows an example of a modulation scheme for the radar sensor according to FIG. 2; [Figure 4] FIG. 10 illustrates a further example of a modulation scheme for a fast chirp FMCW radar according to another embodiment. [Figure 5] FIG. 5 is a simplified diagram of details of a modulation scheme based on the principle shown in FIG. 4. [Figure 6] FIG. 5 shows another variant of the modulation scheme corresponding to FIG. 4. [Figure 7] 1 shows a schematic diagram of a car with a radar sensor and an interfering signal source; DETAILED DESCRIPTION OF THE INVENTION

[0015] A typical configuration of a radar sensor to which the present invention can be applied will be described based on the block diagram shown in Figure 1. The radar sensor comprises a functional group 10 for generating a transmit signal 12 in the radar frequency range (76-81 GHz). This signal is applied to a transmit antenna 16 via a parallel amplifier and phase shifter circuit 14.

[0016] Received signals, e.g., radar echoes received from various radar targets, are received via multiple parallel receive antennas 18 and mixed with the transmitted signal 12 by mixers 20, thereby converting them into so-called baseband signals having lower frequencies. In the baseband range, the signals are amplified, filtered, and digitized by appropriate circuits 22.

[0017] Further processing of the baseband signals to detect radar targets and calculate measurement variables such as range, relative velocity and position angle of the radar targets takes place in digital processing stage 24 .

[0018] The characteristics of a radar sensor with regard to measuring object distance and velocity are essentially determined by the signal shape or modulation of the transmitted signal. Current radar sensors are primarily based on variations of the so-called fast chirp or chirp sequence modulation shown in Figure 2, in which the frequency f of the transmitted signal 12 is expressed as a function of time t. The modulation sequence 26 for detecting the environment is a time-domain modulation of the transmitted signal 12 at equidistant time distances T R2R The FMCW ramps consist of a series of identically shaped FMCW ramps or chirps 28 delivered at .

[0019] The baseband digital received signal is transformed into a spectrum 30 using a two-dimensional Fourier transform (Fast Fourier Transform, FFT) for each transmit and receive channel. In the first dimension D1, a Fourier transform is performed on each individual chirp 28. The input variable is the complex amplitude of the baseband signal at a time corresponding to a series of support points 32 on the chirp 28. Each power peak 34, 36 in this one-dimensional spectrum in dimension D1 corresponds to a radar target at a specific distance from the radar sensor proportional to the frequency of the peak. In the second dimension D2, a Fourier transform is performed on corresponding support points 32 on successive chirps 28. Thus, the reflected signal produces power peaks 34', 36' in the two-dimensional spectrum 30, whose frequency location in dimension D2 indicates the relative velocity of the radar target of interest. Thus, the spectrum 30 is divided into range bins 38 in the first dimension D1 and velocity bins 40 in the second dimension D2. Within each range bin 38, the distribution of signal power across velocity bins 40 represents the Doppler spectrum. If the reflected signal power is sufficient, power peaks 34', 36' assigned to various radar targets can be detected and assigned to respective range and velocity bin pairs. Depending on the modulation parameters of the chirp sequence modulation, each range bin 38 is assigned a specific range of object ranges, and each velocity bin 40 is assigned a specific range of relative velocities. Therefore, the coordinates (separation and velocity) of the located radar target can be determined based on the range bin and velocity bin indexes.

[0020] The position angle of the radar target is determined in a known manner based on the phase relationships between the signals received on the various receive channels. In accordance with the present invention, the above-described method, and in particular the modulation scheme used, is modified to more deliberately suppress signals from known interferers in the spectrum 30. Instead of sending the chirps equidistant in time, the start times of the individual chirps 28 are each sent over a nonlinearly varying time interval ΔT. nAs a result, the sampling instant for Doppler processing of the received signal of the radar sensor is subject to a corresponding "fluctuation" or "jitter." As shown in FIG. 3, the starting instant t n (for n>1) is determined according to the following equation (1): t n =t1+(n-1)T0+ΔT n (1) Therefore, the distance between the start of two consecutive chirps 28 is a function of the fixed component T0 and the variable time interval ΔT n where the variable time interval ΔT n is always greater than 0 and less than a certain fluctuation band Δ. The upper limit on the fluctuation band Δ is imposed by the fact that successive chirps 28 must not overlap each other. Therefore, T r When is the lamp duration, the following must hold:

[0021] Δ≦T0-T r (2) Time interval ΔT n should vary randomly, i.e., in particular non-linearly, and can be generated, for example, using a random number generator. Optionally, a fixed series of time intervals ΔT are repeated after a certain (possibly large) number of chirps 28. n Preferably, the time intervals are evenly distributed over the interval [0, Δ].

[0022] Another variation of the conventional chirp sequence method is shown in Figure 4, where packets P1, P2, ..., P3 are composed of successive chirps 28, each separated by a time gap. n The center frequencies of chirp 28 are f1, f2, ..., f n , P1, P2, ..., P2, ..., P1, ... ... n At the start of τ1, τ2, …, τ n is a constant time distance T p2pare separated from each other by

[0023] A possible adaptation of the present invention to this version of the chirp sequence method is shown in Figure 5, where three consecutive packets P n , P n+1 , P n+2 Only the start sequence of τ is shown. m , τ m+1 , or τ m+1 , τ m+2 The interval between the fixed component T0 and the variable time interval ΔT n+1 or ΔT n+2 In contrast, the distance T between the start times of successive chirps 28 within each packet is R2R is constant. Also, the time interval ΔT n+1 , ... can vary only within a certain variation band Δ, which in this case is limited by the requirement that successive packets must not overlap each other. n The larger the time gap between..., the greater the time interval ΔT n+1 , there is more room for change in...

[0024] Figure 6 shows the packet start time τ n , τ n+1 5, but also varies the start times of the chirps 28 within each packet. Within each packet, the spacing between the start times of the chirps varies with a fixed component T0 and a variable portion ΔT2, ΔT3, or ΔT K , ΔT K+1 where the subscripts k and k+1 indicate that the sequence of varying time intervals may be different for different packets. Optionally, of course, the same variation scheme can be used for all packets.

[0025] Time interval ΔT n The effect of the change in can be explained based on the following signal model: The phase of the received signal resulting from the nth chirp or nth packet due to reflection from a single object

[0026]

number

[0027] With respect to , the following relationship holds:

[0028]

number

[0029] where f n is the center frequency of the chirp or packet, c is the speed of light, v is the relative velocity of the object, the starting time of the nth chirp or nth packet, and

[0030]

number

[0031] is the phase of the received signal that would be obtained for an object with relative velocity v = 0. i When an interference signal having the following relationship holds true:

[0032]

number

[0033] In this equation, (2f n / c) is the Doppler frequency f d which is preferably determined as accurately as possible within the spectrum 30 in order to measure the relative velocity v. That is,

[0034]

number

[0035] Time interval ΔT n The change in the interference signal term 2πf i ΔTn The desired phase modulation of the interference signal by and the effective signal term 2πf d ΔT n These phase modulations result in a "smearing" of the output peaks over a frequency range that may span multiple velocity bins, and a corresponding flattening of the peaks, due to the Fourier transform in the second dimension D2. i where the time interval ΔT n The fluctuation range Δ is Δ=1 / 2f i The interference signal peaks are then broadened and flattened as desired and are therefore no longer misinterpreted as reflections from objects. Furthermore, since the chirp sequence consists of pairs of chirps that are 180° out of phase with each other due to phase modulation, all odd harmonics of the interference signal frequency are suppressed. In contrast, at the useful signal peaks, smearing is suppressed by a factor of 1 / f i <<1, the side lobes of the useful signal peak are effectively suppressed regardless of the object's Doppler frequency, and therefore real radar targets can still be detected well.

[0036] 7, a vehicle 42 is shown that includes an interfering signal source 44 with a known frequency spectrum. The vehicle is further equipped with a radar sensor 46 that operates according to the principles described above. The fluctuation range Δ is determined by the frequency at which the interfering signal from the interfering signal source 44 reaches its main frequency f i , so that the interfering signal is effectively suppressed when evaluating the radar echo.

Claims

1. A method for suppressing an interfering signal having a known interfering signal frequency spectrum in a fast chirp FMCW radar sensor (46) for a motor vehicle (42), comprising: subjecting a baseband signal formed by mixing a transmitted signal with a received signal to at least a two-dimensional Discrete Fourier Transform; one (D2) of at least two dimensions (D1, D2) representing a Doppler spectrum, i.e., a spectrum of Doppler frequencies resulting from the radial velocity of a radar target relative to the radar sensor; and determining whether a chirp (28) begins in the transmitted signal within a time interval ΔT during which the distance of the chirp (28) varies nonlinearly. ｎ wherein the time intervals Δt ｎ The fluctuation range Δ of at least one interference signal frequency f in the Doppler spectrum depends on the interference signal frequency spectrum. ｉ is intentionally selected to be suppressed.

2. The transmitted signal is a series of packets (P １ , ..., P ｎ ), and the packet (P １ , ..., P ｎ ) each containing a plurality of chirps 28, separated from one another by time gaps, and １ , ..., P ｎ ) at the start time τ １ , …, τ ｎ The distance between the time intervals Δt changes nonlinearly ｎ The method of claim 1 , wherein each of the first and second digits differs from the other by only one of the following:

3. The chirps (28) within each packet are spaced apart in time by an equal distance T Ｒ２Ｒ 2. The method of claim 1, comprising:

4. Various packets (P １ , ..., P ｎ ) have their center frequencies (f １ , f ２ 4. The method according to claim 2 or 3, wherein the aryl groups are different from each other in the following:

5. The time interval ΔT ｎ The fluctuation range Δ is 1 / 4f ｉ ~1 / f ｉ The interference signal main frequency f ｉ 5. A method according to claim 1 for suppressing an interfering signal having

6. The chirp (28) or the lamp (P １ , ..., P ｎ ) at the start time (t １ , ..., t ｎ ;τ １ , …, τ ｎ ) are fixed components (T ０ ) and the varying time interval (ΔT １ , ..., ΔT ｎ ) additively composed of successive time intervals (ΔT １ , ..., ΔT ｎ 6. The method of claim 1, wherein the Δ seqs form a random or pseudo-random sequence and are uniformly distributed over the interval [0, Δ].

7. An internal interference signal source is provided, and the interference signal frequency f ｉ 6. A radar sensor, in which the method according to any one of claims 1 to 5 is implemented to suppress

8. 6. A vehicle (42) comprising an internal interference signal source (44) and a radar sensor (46), wherein the method of any one of claims 1 to 5 for suppressing interference signal frequencies of the interference signal source (44) is implemented.

Images