Angle-resolving radar system

Abstract

Angle-resolving radar system with two radar sensors (10, 12) arranged at a given distance (d), configured to transmit coherent radar signals, and oriented such that the signal transmitted by one of the radar sensors and reflected from an object (14) can be received by the other sensor, characterized in that the frequencies f1 and f2 of the signals transmitted by the two radar sensors (10, 12) have a given frequency difference f2-f1, that at least one (12) of the radar sensors is configured to mix a signal transmitted by the other sensor (10) and reflected from the object (14) with the signal transmitted by the other sensor (10), and that an evaluation device (22) is configured to compare the baseband signal (B) resulting from the mixing with a reference signal (R) that is identical to the signal transmitted by the other sensor (10).and to determine a localization angle (φ) of the reflecting object (14) based on the phase difference of these signals.

Description

[0001] The invention relates to an angle-resolving radar system comprising two radar sensors arranged at a given distance, which are configured to transmit coherent radar signals and which are oriented such that the signal transmitted by one of the radar sensors and reflected by an object can be received by the other sensor. State of the art

[0002] Radar systems of the above-mentioned type are called cooperative radar systems and are increasingly being used in motor vehicles to detect the traffic environment.

[0003] State-of-the-art angle-resolving radar systems are described, for example, in: Schoor “High-resolution angle estimation for automotive radar systems”, Verlag Dr. Hut, 2010.

[0004] In angle-resolving SIMO radar systems, the aperture size is limited by the fact that only a limited number of receive channels can be provided with reasonable effort. In MIMO (multiple input, multiple output) radar systems, the attempt is made to arrange the transmit and receive antennas in such a way that, with a given number of transmit and receive channels, a large virtual aperture and thus high angular separation capability is achieved, while the side lobes that inevitably occur with a limited number of antennas due to sometimes large antenna spacing are suppressed as much as possible. However, the goal of an unambiguous and simultaneously high-resolution angle estimation is difficult to achieve with these approaches; for example, a compromise with a reduced dynamic range is required. Disclosure of the invention

[0005] The object of the invention is to improve the angle estimation in a cooperative radar system.

[0006] This object is achieved according to the invention in that the frequencies f1 and f2 of the signals transmitted by the two radar sensors have a given frequency spacing f2-f1, in that at least one of the radar sensors is configured to mix a signal transmitted by the other sensor and reflected by an object with the signal transmitted by the other sensor, and in that an evaluation device is configured to compare the baseband signal produced by the mixing with a reference signal which is identical to the signal transmitted by the other sensor and to determine a location angle of the reflecting object based on the phase difference of these signals.

[0007] In cooperative radar systems, the large spatial separation between the two or more radar sensors results in a very large aperture. However, since these separations are significantly greater than the wavelength of the radar radiation, the Nyquist uniqueness criterion is severely violated. According to the invention, the signals transmitted by the radar sensors are phase-locked to each other, but they have different frequencies. The frequency separation between these signals makes it possible to generate a baseband signal with a significantly lower (beat) frequency f2-f1, which corresponds to a fictitious radar wave with a correspondingly longer wavelength. The signal received from a radar target then exhibits an angle-dependent phase difference compared to the reference signal, which enables angle estimation.Due to the long wavelength of the simulated radar wave, the angle estimation is unambiguous at least over a wide angular range, and due to the large aperture, a high resolution is still achieved.

[0008] The invention is not limited to radar systems with exactly two radar sensors, but is also applicable to cooperative systems with more than two radar sensors. Likewise, the signal transmitted by each individual radar sensor can also be a superposition of signals with multiple frequencies, provided that all frequency components of the two cooperating radar sensors have the same frequency spacing.

[0009] Advantageous embodiments and further developments of the invention emerge from the subclaims.

[0010] The radar sensors participating in the cooperative system can themselves be angle-resolving, meaning they enable angle estimation based solely on the signals transmitted and received by the same sensor. For example, one or more radar sensors in the system can be configured as MIMO radar sensors, which have a high angular resolution due to the short wavelength corresponding to the frequency f1 or f2, but only a very limited unambiguous range due to the large aperture. For an object that can be located by both sensors, the unambiguous angle estimate based on the baseband signal with the frequency f2-f1 can then be used to resolve the ambiguity of the high-resolution angle estimate.

[0011] If the detection range of a single radar sensor overlaps with the detection range of two neighboring sensors, this sensor can cooperate with the neighbor who also detects the object.

[0012] The radar signals transmitted by the various sensors can be derived in a phase-locked manner from a fundamental frequency f0 of a synchronization signal, for example by multiplying f0 and modulating a multi-tone baseband signal, in particular an OFDM (Orthogonal Frequency Division Multiplex) signal.

[0013] The following example is explained in more detail using the drawing. It shows: Fig. 1 A block diagram of a radar system according to the invention; Fig. 2 a sketch to explain the functionality of an angle estimation method implemented in the radar system; and Fig. 3 a motor vehicle with a radar system according to the invention in plan view.

[0014] In Fig. 1 shows a block diagram of two cooperative radar sensors 10, 12 which are installed, for example, in a motor vehicle (not shown) in such a way that their detection areas overlap each other, so that an object 14 present in the overlap area can be detected by both sensors.

[0015] The two radar sensors 10, 12 are configured as MIMO radar sensors and each have a high-frequency part 16 or 18 and a digital evaluation device 20 or 22.

[0016] The radio-frequency section 18 has a MIMO antenna array 24 with transmitting antennas 26 and receiving antennas 28. For reasons of clarity, only two transmitting antennas and three receiving antennas are shown here. The radar signal to be transmitted by the transmitting antennas 26 is generated by a local oscillator 30. Each of the receiving antennas 28 is assigned a mixer 32, which mixes the received signal with the signal generated by the local oscillator 30 (the corresponding signal lines are not shown here), so that a low-frequency intermediate-frequency signal is generated. The intermediate-frequency signals generated in the various receiving channels are digitized and then further evaluated in the evaluation device 22. In a known manner, an ambiguous set of angle estimates φi_12 is generated in an angle estimator 34, each of which indicates a candidate for the azimuth angle of the object 14.The ambiguity is resolved in a further evaluation stage 36, as will be explained in more detail below.

[0017] The radio-frequency section 16 of sensor 10 has a MIMO antenna array 38, which may be identical to the array 24 of the other sensor, but is shown here as a mirror image for clarity. The transmit antennas 40 of array 38 are fed by a modulator 42, which receives a synchronization signal 44 from the local oscillator 30 of the other sensor 12 and generates a transmit signal phase-locked to the transmit signal of sensor 12. For example, the local oscillator 30 generates a signal with a frequency f1 of 76.5 GHz, and the modulator 42 phase-locks this signal into a signal with a frequency f2 of 76.7 GHz, with a fixed frequency spacing f2-f1 = 200 MHz.

[0018] Each receiving antenna 46 of the MIMO array 38 is assigned a mixer 48, which mixes the received signal with the signal generated by the modulator 42. The intermediate frequency signals thus obtained in each receiving channel are digitized and further processed in the evaluation device 20. An angle estimator 50 generates an ambiguous set of angle estimates φi_10, which are transmitted to a further evaluation stage 52.

[0019] A portion of the signal transmitted by the transmitting antennas 46 of sensor 10 can be reflected by the object 14 such that the reflected signal is received by the receiving antennas 28 of sensor 12. At least one of these receiving antennas is capable of forwarding the received signal not only to the associated mixer 32, but also to another mixer 54. Depending on the embodiment, the signal from this antenna can be split and fed to both mixers 32, 54, or the two mixers can be alternately switched. The high-frequency section 18 of sensor 12 contains a modulator 56, which has the same function as the modulator 42 in sensor 10 and thus generates a signal with the frequency f2. In the mixer 54, this signal is mixed with the signal received by the antenna.This generates a baseband signal B whose frequency approximately corresponds to the frequency spacing f2-f1, but depends on the propagation time of the signal from the transmitting antennas 46 via the object 14 to the receiving antennas 28 and thus on the location angle of the object.

[0020] In the evaluation device 22, the baseband signal B is compared with a reference signal R in a bistatic angle estimator 58. This reference signal is generated by a mixer 60, which mixes the output signal of the local oscillator 30 with the output signal of the modulator 56. The reference signal R thus has exactly the frequency f2-f1. As will be shown below with reference to the Fig. 2, a unique value φ_12 for the azimuth angle of the object 14 can be determined from the phase difference between the baseband signal B and the reference signal R. This value is transmitted to the evaluation stage 36, where it allows the selection of the high-resolution but ambiguous estimated values ​​φi_12 that best matches φ_12 and thus indicates the true angle φ*_12 at which the subject 14 is seen by the sensor 12.

[0021] Furthermore, the unique value φ_12 is also transmitted to the evaluation stage 52 in the sensor 10 and used there to select from the high-resolution but ambiguous estimated values ​​φi_10 the one that best agrees with φ_12 and indicates the true angle φ*_10 at which the object 14 is seen by the sensor 10.

[0022] The evaluation devices 20 and 22 contain further evaluation stages, which are Fig. 1 and serve to determine the distance and relative speed of the object 14. For the angle estimation method proposed here, it is expedient if the signals transmitted by the transmitting antennas 26 and 40 have a fixed frequency or consist of a superposition of signals with fixed frequencies, as is the case, for example, with an OFDM radar. With this type of radar, the transmitted signal is a superposition of orthogonal subcarrier frequencies that do not change over time. The waveforms of the transmitted signal are digitally modulated and form a sequence of code symbols that are digitally demodulated at the receiver end. Thus, unlike, for example, with an FMCW radar, the distance and angle information is obtained here not by temporal variation of the frequency, but by temporal variation of the code symbols.

[0023] In Fig. Figure 2 schematically shows the two sensors 10, 12, which are spaced apart by a known distance d of, for example, 75 cm. The distance to object 14 is much greater than the distance d, so that the lines of sight S10 and S12 from sensor 10 and from sensor 12 to object 14, respectively, are almost parallel. Line of sight S10 forms an angle φ with the optical axis A of sensor 10, which corresponds to the azimuth angle of object 14. Line of sight S12 forms approximately the same angle with the optical axis of sensor 12 (not shown here).

[0024] In Fig. 2 shows two wavefronts W1 and W2 of a fictitious wave that propagates at the speed of light c and has the frequency f2-f1 and thus the wavelength λ = c / (f2-f1). The wavefront W1 has just reached the sensor 12 and will reach the sensor 10 with a time delay Δt that depends on the detection angle φ to the object 14 and is proportional to the path length difference c Δt = d sin(φ). The phase difference δ between the baseband signal B and the reference signal R is related to 2π as the path length difference c Δt is to the wavelength λ. Thus, the detection angle φ can be determined from the measured phase difference. The estimated value for the detection angle thus obtained is unambiguous as long as the distance d = max{c Δt} is smaller than c / (2 (f2-f1)).

[0025] In Fig.Figure 3 shows a top view of a motor vehicle 60 equipped with a cooperative radar system. The radar system comprises a total of six radar sensors 10, 12, and 62 to 68, which are symbolically represented here in the form of stylized radar beams. The radar sensors are, for example, OFDM radars that transmit a sequence of code symbols with fixed subcarrier frequencies within each measurement cycle. The sequences of transmitted signals are interrupted by transmission pauses during which an electronic evaluation of the received radar echoes takes place. The measurement cycles of all six radar sensors are synchronized with each other. The radar sensors of each pair of adjacent sensors can cooperate with each other. For example, sensor 12 can cooperate with sensor 10 or with sensor 62. Whenever two sensors locate the same object, they can cooperate with each other and make a precise angle estimate in the manner described above.If a single object is located in front of three or more radar sensors in the detection range, the angle estimation method can be further refined by evaluating the phase differences and distances of these multiple sensors.

Claims

[1] Angle-resolving radar system comprising two radar sensors (10, 12) arranged at a given distance (d) which are configured to transmit coherent radar signals and which are oriented such that the signal transmitted by one of the radar sensors and reflected by an object (14) can be received by the other sensor, characterized bythat the frequencies f1 and f2 of the signals transmitted by the two radar sensors (10, 12) have a given frequency spacing f2-f1, that at least one (12) of the radar sensors is configured to mix a signal transmitted by the other sensor (10) and reflected by the object (14) with the signal transmitted by the other sensor (10), and that an evaluation device (22) is configured to compare the baseband signal (B) produced by the mixing with a reference signal (R) which is identical to the signal transmitted by the other sensor (10), and to determine a location angle (φ) of the reflecting object (14) on the basis of the phase difference between these signals. [2] Radar system according to claim 1, comprising at least three radar sensors (10, 12, 62...68) which cooperate with each other when they locate the same object (14). [3] Radar system according to claim 1 or 2, wherein at least one of the radar sensors (10, 12) is angle-resolving on its own. [4] Radar system according to claim 3, wherein the at least one angle-resolving radar sensor (10, 12) is a MIMO radar. [5] Radar system according to claim 3 or 4, wherein the at least one angle-resolving radar sensor (10, 12) is configured for a high-resolution but ambiguous angle estimation and the location angle (φ) obtained by the cooperation of the radar sensors serves to resolve the ambiguity. [6] Radar system according to one of the preceding claims, wherein at least one of the radar sensors is an OFDM radar.

Images