Radar sensor network and method for determining the relative velocity of a radar target

Abstract

A radar sensor network (100) and method for determining the relative velocity (v) of a radar target (18), wherein a plurality of transmitting antenna elements (14) of a plurality of radar sensors (10) transmit phase-modulated transmission signals (30-1; 30-2; 30-3), and each radar sensor's ramp-like frequency-modulated transmission signals (30-1; 30-2; 30-3) have a sequence (1a; 1b; 1c; 1d; ...) of ramps (40) that are temporally interleaved with each other, and these ramps (40) successively with a time staggeration of a predetermined time interval (Tr2r) within each sequence (1a; 1b; 1c; 1d; ...) The radar sensor network (100) and method include a radar sensor network (100) and method in which the phase of the transmitted signals (30-1; 30-2; 30-3) is phase-modulated with a code for each transmitting antenna element (14) of the radar sensor (10), and the measurement cycles of the multiple radar sensors (10) include temporally sequential intervals (42), each of which is assigned to one of the radar sensors (10), and includes one or more ramps (40) of the transmitted signals (30-1; 30-2; 30-3) of only the radar sensor (10) assigned to this interval; from the received response signals, an estimate of the relative velocity (v) of the radar target (18) is determined.

Description

[Technical Field]

[0001] The present invention relates to a radar sensor network and method for determining the relative velocity of a radar target. This radar sensor network may, in particular, be a radar sensor network for automobiles. The radar sensors of this radar sensor network may be adapted, in particular, for FMCW (Frequency Modulated Continuous Wave) measurement methods using linear frequency ramps, and for evaluation of received signals using discrete Fourier transforms, in particular FFT (Fast Fourier Transform). [Background technology]

[0002] Radar sensors are used to implement comfort features such as adaptive cruise control and safety features such as emergency brake assist. The essential advantage of such sensors is that they directly measure physical quantities and do not involve the interpretation of images, such as video camera images. A radar sensor transmits a high-frequency radar beam through an antenna structure and receives the beam reflected by an object. In this regard, the object being captured may be stationary or moving. Using the received radar beam, the distance and direction (angle) to the object can be calculated. In addition, the relative velocity of the object to the radar sensor can be calculated. Typical radar sensors operate in the frequency range between 76 and 81 GHz.

[0003] Regarding radar modulation, the so-called chirp sequence method is a known technique. In this case, many high-speed frequency ramps, or chirps, are transmitted. To improve angular accuracy and separation, multiple transmitting antennas are often used to achieve an expanded virtual aperture via the MIMO (Multiple-Input-Multiple-Output) principle. The transmitted signals must be separated upon reception. A simple method for this is time-division multiplexing, in which case the time interval of chirps from the same transmitting antenna increases by a coefficient N = the number of transmitting antennas. As a result, the unique measurement range of relative velocity also decreases by a coefficient N.

[0004] U.S. Patent Application Publication No. 2019 / 0353770(A1) and DE102017200317A1 describe a radar sensor and method for determining the relative velocity of a radar target. The velocity estimation of an object is performed using a radar sensor with multiple transmitting antennas. Multiple interleaved frequency ramp sequences are transmitted using the multiple transmitting antennas. In this regard, individual phase coding using harmonic codes is performed for each transmitting antenna. To estimate the velocity of an object, ambiguity based on code division multiplexing is eliminated.

[0005] Disclosure of the invention In cooperative radar sensor networks, systems with coherent distribution of HF oscillator signals and systems without distribution of HF oscillator signals are distinguished. Systems with distribution of HF oscillator signals clearly exhibit better phase noise behavior; however, since the distribution of oscillator signals is associated with high costs, solutions without distribution of oscillator signals are more advantageous for practical implementation.

[0006] In the case of a cooperative radar sensor network without HF oscillator signal distribution, evaluation of the response signal may include evaluating the response signal of one radar sensor to a transmitted signal from another radar sensor, received within one radar sensor of the radar network. For this purpose, calibration or correction between non-coherent radar sensors may be performed. The evaluation may be performed based on techniques described in, for example, the following two documents: M. Gottinger, F. Kirsch, P. Gulden, and M. Vossiek, "Coherent Full-Duplex Double-Sided Two-Way Ranging and Velocity Measurement Between Separate Incoherent Radio Units," IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 5, pp. 2045-2061, May 2019, doi:10.1109 / TMTT.2019.2902553. A. Duerr, B. Schweizer, J. Bechter, and C. Waldschmidt, "Phase Noise Mitigation for Multistatic FMCW Radar Sensor Networks Using Carrier Transmission," IEEE Microwave and Wireless Components Letters, vol. 28, no. 12, pp. 1143-1145, December 2018, doi:10.1109 / LMWC.2018.2878983.

[0007] When the code division multiplexing method of U.S. Patent Application Publication No. 2019 / 0353770(A1) and DE102017200317A1 is used for multiplexing multiple sensors in a cooperative sensor network without a common oscillator signal, the sensitivity and dynamics of the radar sensor network will be severely limited by uncorrelated phase noise from different local oscillators, which generally causes a strong increase in noise levels in the received signals of all sensors. Monostatic signals (transmission and reception on the same radar sensor using its own oscillator signal) will also suffer from this degradation. [Overview of the project] [Problems that the invention aims to solve]

[0008] Therefore, the object of the present invention is to present a modulation method for radar sensor networks that enables unambiguous velocity measurement within a single sensor while simultaneously achieving good phase noise behavior in the monostatic signal response. [Means for solving the problem]

[0009] This problem is solved by the present invention, which provides a method for determining the relative velocity of a radar target. The step of transmitting a phase-modulated transmission signal using multiple transmitting antenna elements of each of the multiple radar sensors, wherein a ramp-like frequency-modulated transmission signal is generated for each of these radar sensors, and this transmission signal has a sequence of multiple time-interleaved ramps, with the ramps in each sequence succeeding at predetermined time intervals, and the phase of this generated ramp-like frequency-modulated transmission signal is phase-modulated with a code for each transmitting antenna element of the radar sensor, and the measurement cycle of the multiple radar sensors includes time-sequential intervals, each of which is assigned to one of the radar sensors, and includes one or more ramps of sensor signals only for the radar sensor assigned to this interval, Steps include receiving a response signal to a transmitted phase-modulated transmission signal, This step involves calculating a two-dimensional spectrum for each sequence of the transmitted signal from each radar sensor by performing a two-dimensional Fourier transform of the baseband signal of the received response signal, wherein the transformation is performed per ramp in the first dimension and along the ramp index of the ramp sequence in the second dimension. A step of determining the periodic relative velocity value of the radar target at a predetermined velocity period, based on a peak in one of the calculated two-dimensional spectra. This is resolved by a method that includes the step of determining an estimate of the relative velocity of a radar target based on the agreement between the phase relationships between two-dimensional spectral values ​​at the same location and the expected phase relationships for multiple determined relative velocity values.

[0010] A ramp sequence is also referred to as a ramp sequence below. Unless otherwise specified, the term "sensor" below refers to a radar sensor. A ramp is also called a chirp. In this configuration, a unique speed measurement is made possible based on a phase-modulated transmission signal transmitted by the transmitting antenna element of one radar sensor and a response signal received (for example, particularly by at least one receiving antenna element of the same radar sensor), without this speed measurement being interfered with by simultaneously transmitted signals from other radar sensors, because at each transmission point, only one of these radar sensors transmits a ramp-like frequency-modulated transmission signal.

[0011] Therefore, the core of the present invention is a modulation method that operates each single sensor using code division multiplexing (e.g., DDM, Doppler division multiplexing), particularly the JSFMCW modulation method (Joint-Sampling-FMCW), and operates different single sensors using time division multiplexing. Since only one sensor transmits a radar beam at any given time, the phase noise behavior of a single sensor is determined solely by the phase noise of the local oscillator or transmit signal processing of that single sensor, thus enabling high sensitivity and signal dynamics in a monostatic response (i.e., the same radar sensor transmits and receives). By using the JSFMCW method with DDM, high sensitivity is achieved within a single sensor (all transmitters of one radar sensor are active simultaneously), and at the same time, unique velocity measurement becomes possible without presupposing the existence of bistatic paths (i.e., paths between different radar sensors).

[0012] This method extends the MIMO principle to a network of multiple cooperative radar sensors, enabling further performance improvements while still addressing the fundamental challenges of unambiguous relative velocity measurement range and transmitter multiplexing.

[0013] In this specification, the terms “monostatic” and “bistatic” are used, on the one hand, for an antenna system of a single radar sensor, in which case a bistatic antenna system has different transmitting antenna elements than receiving antenna elements, and a monostatic antenna concept of a single radar sensor where the same antenna elements are used for both transmitting and receiving. On the other hand, in this specification, the terms “monostatic” and “bistatic” are used for a received radar signal of one of several radar sensors, in which case the term “monostatic” is used to mean that the received radar signal originates from a transmitted signal sent by the same radar sensor, and the term “bistatic” is used to mean that the received radar signal originates from a transmitted signal sent by another of several radar sensors.

[0014] The transmitting and receiving antenna elements of a radar sensor can be formed similarly. These may, for example, each consist of a patch antenna array. Receiving a response signal to a transmitted phase-modulated transmit signal may include receiving at least a response signal by at least one receiving antenna element of the same radar sensor for a phase-modulated transmit signal transmitted by the transmitting antenna element of that radar sensor. In other words, at least a monostatic response signal is received.

[0015] The measurement cycles of multiple radar sensors include temporally consecutive intervals, each of which is assigned to one of the radar sensors and includes one or more ramps of the transmitted signal of the radar sensor assigned to that interval. In particular, each of these intervals may include one or more ramps of the transmitted signal of only the radar sensor assigned to that interval. Thus, at each point in time, the transmitted phase-modulated ramp-like frequency-modulated transmitted signal of at most one of the multiple radar sensors travels through one ramp.

[0016] In other words, each radar sensor transmits at a time staggered based on time-division multiplexing. To put it another way, each individual ramp in the sequence of one of these radar sensors does not overlap in time with the ramps in the sequence of another of these radar sensors, or is arranged in a way that the time-division multiplexing of the radar sensors prevents them from overlapping in time with the ramps in the sequence of another of these radar sensors. In particular, while one ramp of the transmission signal of each radar sensor is passing through, only that radar sensor is transmitting. Therefore, at each point in time, the transmission signal of only one of the radar sensors (i.e., the transmission signal of only one of the radar sensor's transmitting antenna elements) is transmitted in a ramp-like frequency-modulated manner. At this time, this phase-modulated transmission signal of the radar sensor is simultaneously transmitted by multiple transmitting antenna elements of that radar sensor. In other words, within each radar sensor, the transmission signal is transmitted simultaneously (synchronously) by multiple transmitting antenna elements, each encoded by phase modulation. A different code is used for each transmitting antenna element that transmits simultaneously. The ramps in the ramp sequence of the transmit signal of one radar antenna element of each of the multiple radar sensors are arranged in the temporal gaps between the ramps in the ramp sequence of the transmit signal of another radar antenna element of each of the multiple radar sensors, i.e., arranged based on a time-division multiplexing scheme. In particular, the transmission of this phase-modulated transmit signal may involve each radar sensor transmitting a phase-modulated ramp-like frequency-modulated transmit signal based on a time-division multiplexing scheme of the multiple radar sensors.

[0017] This method may include the following steps: assigning a single response signal to one of several transmitting antenna elements. This can be done, in particular, based on the code of the transmitting signal of the transmitting antenna element. Possible codes for assigning the response or transmitting signal to a single transmitting antenna element include, for example, noise or harmonic codes. Based on this assignment, angle estimation of the detected radar target can be performed.

[0018] In this embodiment, the code is a harmonic code. Harmony, in this context, means that the phase modulation represents discrete harmonic oscillations. Thus, a harmonic code here means that the phase of the generated ramp-like frequency-modulated transmitted signal is phase-modulated by one (complex) harmonic code for each transmitting antenna element of a radar sensor, and this harmonic code represents discrete harmonic oscillations. Possible harmonic codes include, for example, a harmonic sequence, e.g., 1, 1, 1, ...; 1, -1, 1, -1, ...; or 1, i, -1, -i, 1, ..., or further harmonic frequencies. Phase modulation by a harmonic code introduces additional ambiguity to the velocity to be estimated. These ambiguities can also be resolved.

[0019] The transmit signals of radar sensors can be generated, for example, from an HF oscillator signal. The transmit signals of multiple radar sensors can be coherent or non-coherent with respect to each other. In some embodiments, the transmit signal and / or HF oscillator signal of each radar sensor are generated by each radar sensor. For example, a reference signal can be generated and distributed to at least one or more of the multiple radar sensors, in which case at least one or more of the multiple radar sensors generate their respective transmit signals based on this reference signal. The reference signal can be, for example, a clock signal or an HF oscillator signal. The latter enables coherent transmit signals from these radar sensors. In other embodiments, the transmit signals of radar sensors are generated by a single HF oscillator and distributed to at least one of the multiple radar sensors. This enables coherent transmit signals from these radar sensors.

[0020] In particular, the calculation of the two-dimensional spectrum for each sequence of the transmission signals of each radar sensor may include calculating the two-dimensional spectrum for each sequence of the transmission signals of each radar sensor for each receiving channel of each radar sensor. That is, for each receiving channel and / or for each receiving antenna element of one radar sensor, the two-dimensional spectrum is calculated for each sequence of the transmission signals of this radar sensor. Therefore, the received response signal includes the radar signal received monostatically. Each receiving channel is assigned to one receiving antenna element.

[0021] In the two-dimensional Fourier transform of the baseband signal of the received response signal, the transform is performed for each ramp in the first dimension and along the ramp index of the ramp sequence in the second dimension. In other words, the transform is performed for each ramp or ramp in the first dimension and along the ramp index that counts the ramps in the sequence in the second dimension.

[0022] Based on the peak in one of the calculated two-dimensional spectra, the value of the relative speed that is periodic with a predetermined speed period of the radar target is determined. In other words, this method includes the step of determining the value of the relative speed that is periodic with a predetermined speed period starting from the peak in one of the calculated two-dimensional spectra of the radar target.

[0023] This method includes the step of determining an estimate of the relative velocity of a radar target based on the agreement between the phase relationships between two-dimensional spectral values ​​at the same location and the expected phase relationships for multiple determined relative velocity values. For example, this method may include identifying agreements between the phase relationships between two-dimensional spectral values ​​at the same location and the expected phase relationships for multiple determined relative velocity values, and selecting an estimate of the relative velocity of the radar target based on the identified agreements in phase relationships. In particular, determining an estimate of the relative velocity of a radar target may include identifying agreements between the phase relationships between two-dimensional spectral values ​​at the same location and the expected phase relationships for multiple determined relative velocity values, and selecting an estimate of the relative velocity of the radar target based on the identified agreements in phase relationships.

[0024] In the embodiment, each of the multiple radar sensors has its own substrate (printed circuit board), and these substrates of the radar sensors are separated from each other. In the embodiment, the multiple radar sensors are arranged in the vehicle at different positions, separated from each other. In particular, the multiple radar sensors have overlapping fields of view or the same field of view.

[0025] The ramp sequences of individual sensors may have the same parameters (chirp duration and frequency deviation, number of chirps, center frequency (which may vary over time), and start time of each chirp), however this is not a prerequisite for this method. In embodiments, within one sequence of ramps, these ramps have the same ramp gradient, the same ramp center frequency difference, the same ramp center frequency, the same ramp duration, and / or the same ramp frequency deviation. In embodiments, ramps with the same ramp index within a ramp sequence of radar sensors each have the same ramp gradient, the same ramp center frequency, the same ramp duration, the same ramp frequency deviation, and / or the same time shift relative to a ramp with one smaller ramp index in the same ramp sequence.

[0026] In this embodiment, the transmitted signals of multiple radar sensors have different time differences between their ramp sequences (more precisely, between the ramps in which each of the individual ramps belonging to each ramp sequence is successive, or between the start times of those ramp sequences), that is, at least one or more time differences between ramp sequences of one radar sensor are different from each of the time differences between the successive start times of ramp sequences of at least one radar sensor of the other radar sensors, or each of the time differences between the successive start times of ramp sequences of each of the other radar sensors.

[0027] In this embodiment, the transmitted signals from multiple radar sensors have different time intervals between the lamps in each lamp sequence, and / or different lamp durations (chirp durations).

[0028] In one embodiment, the lamp sequences of each radar sensor have equal time differences relative to each other. In other words, the lamps with the same lamp index in each radar sensor's lamp sequence have equal time differences relative to each other. That is, their lamp start times are arranged at equal time intervals. In another embodiment, the lamp sequences of each radar sensor are paired and have unequal time differences relative to each other. In other words, the lamps with the same lamp index in each radar sensor's lamp sequence are paired and have unequal time differences relative to each other. That is, their lamp start times are arranged at unequal time intervals.

[0029] In the embodiment, the transmitted signal of each radar sensor includes or constitutes a sequence of lamps arranged in temporal order, and each lamp group includes (or is within the scope of) one lamp from each of the lamp sequences of the transmitted signal of the radar sensor. These may, in particular, be lamps having the same lamp index. Thus, the lamps of the lamp group sequence are temporally consecutive. This succession may be seamless and / or spaced apart.

[0030] In the embodiment, the measurement cycles of multiple radar sensors include temporally sequential blocks, each block including one group of lamps from each of the radar sensors. In the preferred embodiment, each block includes exactly one group of lamps from each of the radar sensors.

[0031] For example, the measurement cycles of multiple radar sensors can be divided into blocks. In particular, the measurement cycles of multiple radar sensors may include temporally consecutive blocks, each block containing (exactly) one group of lamps from each of the radar sensors, and the lamps of one group of lamps in this block are temporally interleaved with the lamps of each other group of lamps in this block.

[0032] In a preferred embodiment, within each block, the lamps of one lamp group in this block are temporally interleaved with the lamps of each other lamp group in this block. Such a radar modulation scheme will hereafter be referred to as radar modulation scheme #1. That is, each lamp group in a block is assigned to at least two segments allocated to the radar sensors of these lamp groups. Thus, the lamps of one lamp group in one radar sensor are interleaved with the lamps of each other radar sensor. More precisely, the lamps of one lamp group in the lamp group sequence of the transmitted signal of each radar sensor are (temporarily) interleaved with the lamps of one lamp group in the lamp group sequence of the transmitted signal of each of the other radar sensors.

[0033] In particular, at least one lamp in each of the lamp sequences of the transmitted signals of another radar sensor can be placed, or passed through, between each of the two successive lamps in each of the lamp sequences of the transmitted signals of one radar sensor.

[0034] In radar modulation scheme #1, multiple radar sensors operate using interleaved time-division multiplexing, which helps to minimize the time lag between radar sensors. In this configuration, multiple transmitting antenna elements are simultaneously active by code division multiplexing, particularly Doppler division multiplexing (DDM).

[0035] The time lags in the ramp sequences of individual sensors can be selected to be equally spaced or unequal, depending on the application and further signal processing. Equally spaced time lags may simplify implementation and optimization in some cases. The detailed form of this modulation scheme can be optimized by considering the characteristics of each ramp sequence in terms of velocity unambiguity. For example, the time interval from chirp to chirp within a single sequence can be varied.

[0036] In embodiments, the method includes the step of performing digital beamforming for a radar target via transmit channels assigned to the transmit antenna elements and / or receive channels assigned to the receive antenna elements of at least two radar sensors of a plurality of radar sensors, wherein when performing digital beamforming, a phase difference between the radar sensors, depending on the relative velocity of the radar target, is taken into consideration. In particular, for a radar target, the respective phase difference between two of the radar sensors can be determined based on at least one (particularly determined) estimate of the relative velocity of the radar target and based on the time difference (within one block) between the ramp groups of the two radar sensors. (Digital) beamforming can also be called (digital) beam shaping. In particular, digital beamforming can be performed via multiple (e.g., all) transmit channels and / or multiple (e.g., all) receive channels of at least two radar sensors. Beamforming via multiple (e.g., all) transmit channels is called transmit-side beamforming. Beamforming via multiple (e.g., all) receive channels is called receive-side beamforming. This beamforming is also called MIMO (Multiple-Input-Multiple-Output) beamforming. The phase difference corresponds to the phase change of the received radar signal, based on the relative movement of the radar target during the time lag between the lamp arrays of different radar sensors. Considering each phase difference makes it possible to coherently evaluate the two-dimensional spectra originating from different radar sensors together. The implementation of digital beamforming may, among other things, involve the appropriate phase summation of the calculated two-dimensional spectra, or the portion of the calculated two-dimensional spectra corresponding to the radar target.

[0037] Digital beamforming can be performed on received monostatic radar signals (response signals), or on received monostatic and bistatic radar signals (response signals). When processing monostatic response signals from different radar sensors, the phase difference between the radar sensors is taken into consideration. When processing bistatic response signals to a phase-modulated transmitted signal received by one radar sensor and transmitted by another radar sensor, the phase difference between the radar sensors is taken into consideration.

[0038] In other words, velocity-dependent phase correction (phase difference) can be determined using a unique velocity measurement of a single sensor to compensate for the time difference between bistatic and / or monostatic signals, thus enabling the coherent processing of signals from multiple sensors together.

[0039] The unambiguous velocity measurement of a single sensor already enables partial processing on the single sensor, for example, MIMO beamforming (transmitter and receiver beamforming) can already be performed on the single sensor. In this case, unambiguous velocity determination and beamforming before or after detection of a radar target can be performed.

[0040] The aforementioned processing can be performed entirely on a central evaluation unit, such as a central control unit, or in a partial manner on a single sensor, such as on an evaluation unit for each individual radar sensor.

[0041] In some embodiments, this method includes the step of determining the angle estimate of a radar target. Determining the angle estimate of a radar target may include, for example, performing digital beamforming for the radar target.

[0042] In embodiments, the implementation of digital beamforming for a radar target includes a first partial step in which the digital beamforming for the radar target is performed via transmit channels assigned to the transmit antenna elements and / or receive channels assigned to the receive antenna elements of at least two radar sensors of a plurality of radar sensors, and a second partial step in which the digital beamforming of each of these at least two radar sensors is coupled, taking into account a phase difference between these radar sensors that depends on the relative velocity of the radar target.

[0043] In the embodiment, the reception of a response signal to a transmitted phase-modulated transmit signal includes the following: for a phase-modulated transmit signal transmitted by the transmitting antenna element of each radar sensor, the response signal is received by at least one receiving antenna element of at least the same radar sensor, and the response signal is received by at least one receiving antenna element of another radar sensor among the plurality of radar sensors. In other words, monostatic and bistatic response signals are received. Thus, evaluation of the response signal may include evaluation of the response signal to a transmit signal of another radar sensor, received within one radar sensor among the radar sensors of the radar network.

[0044] If necessary (for example, due to excessively high phase noise in the bistatic response), the evaluation of the bistatic signal can be abandoned instead, and only the monostatic signals of individual sensors can be evaluated. This allows for alternative solutions without the need to change the modulation method.

[0045] In one embodiment, the steps include calculating a two-dimensional spectrum, determining a value for the relative velocity of a radar target, determining an estimate of the relative velocity of the radar target, and optionally performing digital beamforming and / or determining an angle estimate for at least the monostatic response signal of the radar sensor, and these steps are optionally also performed for the bistatic response signal of the radar sensor. This method may include a step of determining whether to perform these steps for the bistatic response signal depending on the (peak) signal-to-noise ratio and / or the noise level of the bistatic response signal.

[0046] In a further preferred embodiment, the measurement cycles of multiple radar sensors include temporally consecutive subcycles, each subcycle being assigned to one of the radar sensors and including the lamp group of the assigned radar sensor. In particular, each subcycle can be assigned to one of the radar sensors and may include the lamp group of only the assigned radar sensor. Such a radar modulation scheme will hereafter be referred to as radar modulation scheme #2. In particular, for example, the lamp group sequences of these radar sensors may be temporally consecutive. That is, the lamp group sequence of the first radar sensor of these radar sensors is followed by the lamp group sequence of the next radar sensor of these radar sensors. Thus, one subcycle assigned to one radar sensor corresponds to one interval assigned to that radar sensor.

[0047] For example, a measurement cycle of multiple radar sensors (a cycle of transmission signals from multiple radar sensors) can include subcycles, within which only one radar sensor transmits, these subcycles occur in succession, and within each subcycle, a time-interleaved ramp sequence of exclusively the single radar sensor in question runs through.

[0048] In radar modulation scheme #2, multiple radar sensors operate sequentially using time-division multiplexing, in which case only one radar sensor transmits within each subcycle. Unlike radar modulation scheme #1, radar modulation scheme #2 does not involve time interleaving of the transmission operations of these sensors. Instead, the individual sensors transmit sequentially, and it is preferable that all sensors receive continuously in this regard.

[0049] In the sequential transmission operation of individual sensors, the accuracy of a single sensor's velocity measurement may, under certain circumstances, not be sufficient to enable coherent processing of the transmitted signals from all sensors. If the accuracy is sufficient, the evaluation can be performed as described above. Otherwise, separate evaluations of "partial cycles" are provided, in which case the transmitted signals of the currently active sensors and the received signals (RX) of all sensors can be evaluated. Because processing is performed in each "partial cycle" due to the sequential transmission operation of individual sensors, the update frequency of detected targets increases proportionally to the number of sensors.

[0050] In evaluating bistatic response signals, the varying activation of the transmitter (TX) during these "partial cycles" results in a variety of virtual antenna arrays with different ambiguous characteristics during angle estimation.

[0051] Instead of separate processing of "partial cycles" and the resulting high update frequency, non-coherent averaging of the results of the partial cycles may also be advantageous. This non-coherent averaging can be performed either before target detection on the spectral plane or after target detection by averaging the angular spectrum.

[0052] In embodiments, the method includes the following steps: performing digital beamforming for a radar target and for applicable partial cycles via a transmit channel assigned to the transmit antenna element of each radar sensor and / or a receive channel assigned to the receive antenna element of each of the multiple radar sensors.

[0053] Digital beamforming can be performed on received monostatic radar signals, or on received monostatic and bistatic radar signals. When processing a bistatic response signal to a phase-modulated transmitted signal received by one radar sensor and transmitted by another radar sensor, the varying activation of the transmitter (TX) in these "partial cycles" results in various virtual antenna arrays with different ambiguities in angle estimation. These ambiguities can be resolved.

[0054] In some embodiments, this method includes the step of determining the angle estimate of a radar target. Determining the angle estimate of a radar target may include, for example, performing digital beamforming for the radar target.

[0055] The determination of the angle estimation may include, namely, determining the ambiguous angle estimations of the radar target for the monostatic and bistatic received response signals of each radar sensor, and determining the unambiguous angle estimation of the radar target based on the ambiguous angle estimations of the radar target.

[0056] In embodiments, digital beamforming is performed for a radar target and for each applicable partial cycle via transmit channels assigned to the transmit antenna elements and / or receive channels assigned to the receive antenna elements of at least two radar sensors of a plurality of radar sensors, the method further includes the step of combining the digital beamforming of at least two radar sensors by non-coherent averaging of the digital beamforming of at least two radar sensors and the applicable partial cycle, or by averaging the angular spectra of the radar target obtained by the digital beamforming of at least two radar sensors and the applicable partial cycle.

[0057] In the embodiment, the reception of a response signal to a transmitted phase-modulated transmission signal includes the following: for a phase-modulated transmission signal transmitted by the transmitting antenna element of each radar sensor, at least a response signal is received by at least one receiving antenna element of the same radar sensor, and the response signal is also received by at least one receiving antenna element of another radar sensor among the plurality of radar sensors. In other words, monostatic and bistatic response signals are received.

[0058] In one embodiment, the problem is solved by a radar sensor network including multiple radar sensors and at least one evaluation unit for determining the relative velocity of a radar target. Each of the multiple radar sensors, Multiple transmitting antenna elements, A signal generator designed to generate a ramp-like frequency-modulated transmission signal and provide it to a transmitting antenna element, wherein the transmission signal has a sequence of multiple time-interleaved ramps, with the ramps in each sequence successively staggered by a predetermined time interval (Tr2r), and the phase of this generated ramp-like frequency-modulated transmission signal is phase-modulated by a code for each transmitting antenna element of a radar sensor, and It includes at least one receiving antenna element designed to receive a response signal to a transmitted phase-modulated transmit signal, The evaluation unit is designed to determine a periodic relative velocity value of a radar target based on a peak in one of the two-dimensional spectra calculated for each sequence of transmitted signals from each radar sensor, and to determine an estimate of the relative velocity of this radar target based on the agreement between the phase relationships between the two-dimensional spectral values ​​at the same location and the expected phase relationships for multiple determined relative velocity values.

[0059] In a preferred embodiment, each of the multiple radar sensors includes multiple receiving antenna elements. Preferably, the multiple receiving antenna elements of a single radar sensor are spatially separated. This allows the radar sensor to constitute a so-called MIMO (multiple input multiple output) system with multiple transmit and receive channels.

[0060] Radar sensor networks are, in particular, collaborative radar sensor networks. The evaluation unit may include a central evaluation unit and local evaluation units for each radar sensor. The local evaluation units may be components of each radar sensor. The evaluation unit may be designed, in particular, to perform the method described above. The radar sensor network may include a control unit. The control unit may include evaluation units. The control unit may be designed to perform the method described above. The control unit may include a central control unit and local control units for each radar sensor. The local control units may be components of each radar sensor. Each local control unit may include a local evaluation unit.

[0061] The following describes exemplary embodiments in more detail with reference to the drawings. [Brief explanation of the drawing]

[0062] [Figure 1] This is a schematic diagram of a cooperative radar sensor network equipped with multiple radar sensors and a central control unit. [Figure 2] This is a schematic diagram of a time-frequency graph of the transmitted signals from multiple radar sensors based on one embodiment. [Figure 3] This is a schematic flowchart illustrating the basis of a method based on one embodiment. [Figure 4] This is a schematic diagram of a time-frequency graph of the transmitted signals from multiple radar sensors based on a further embodiment. [Figure 5] This is a schematic diagram of a virtual antenna array corresponding to another embodiment. [Figure 6] This is a schematic flowchart of a method based on a further embodiment. [Modes for carrying out the invention]

[0063] Figure 1 shows a schematic diagram of a radar sensor network 100 comprising multiple FMCW radar sensors 10-1 and 10-2 and a central control unit 20. In reality, the number of radar sensors 10 can be much larger. Each of the radar sensors 10-1 and 10-2 includes a local control and evaluation unit 12. Each of the sensors 10-1 and 10-2 further includes two transmitting antenna elements 14-1 and 14-2 and two receiving antenna elements 16-1 and 16-2. In reality, many more antenna elements are possible. The radar sensors 10-1 and 10-2 are mounted on the same side of the vehicle, for example, the front, and are adapted to measure the distance d, angle, and relative speed v of a radar target 18, for example, a vehicle or obstacle traveling ahead.

[0064] In the example shown, radar sensors 10-1 and 10-2 each have a bistatic antenna system that uses different antenna elements for transmission and reception. However, a monostatic antenna system in which each radar sensor uses the same antenna element for transmission and reception can also be used.

[0065] Each control and evaluation unit 12 includes a signal generator 17 that generates a transmit signal, which is designed to individually modulate the phase of the transmitted signal supplied to each transmitting antenna element 14-1, 14-2. In this example, harmonic codes are used for the phase modulation of the transmit signal, and different codes are used for each transmitting antenna element 14-1, 14-2. The phase-modulated transmit signal is supplied to multiple transmitting antenna elements 14-1, 14-2 and transmitted by these transmitting antenna elements 14-1, 14-2. The transmitted radar signal, reflected by object 18, is received by receiving antenna elements 16-1, 16-2. The received signal is down-converted to a baseband signal and evaluated by the control and evaluation unit 12.

[0066] The frequency of the transmitted signal can be modulated by a sequence of rising or falling ramps within the measurement cycle of the radar measurement. An example of a ramp-like frequency-modulated transmitted signal is described in more detail below.

[0067] Antenna elements 14-1, 14-2, 16-1, and 16-2 can be positioned at different locations in the direction of angular resolution of the radar sensor 10. This requires, in particular, multiple receiving antenna elements 16-1 and 16-2 arranged in a straight line at uniform intervals (ULA; Uniform Linear Array). The same applies to transmitting antenna elements 14-1 and 14-2, in which case the transmitting and receiving antenna elements 14-1, 14-2, 16-1, and 16-2 do not necessarily have to be arranged on the same straight line. When the radar sensor 10 is used to measure the azimuth angle of an object, the straight line on which the antenna elements are arranged runs horizontally. For a sensor used to measure the elevation angle, the antenna elements would likely be arranged on a straight line perpendicular to this. A two-dimensional antenna array capable of measuring both azimuth and elevation angles is also conceivable.

[0068] The transmitting antenna elements 14-1, 14-2 and the receiving antenna elements 16-1, 16-2 can each be formed similarly and, for example, have a matching field of view. The transmitting and receiving antenna elements 14-1, 14-2, 16-1, 16-2 can each consist of, for example, patch antenna arrays.

[0069] The central control and evaluation unit 20 controls the radar sensors 10-1 and 10-2 and the evaluation unit 12, and is responsible for the part of the evaluation of the received radar signals that includes the radar sensors.

[0070] Figure 2 schematically shows a graph relating the frequencies f of the transmitted signals 30-1, 30-2, and 30-3 of a cooperative radar sensor network 100, which basically corresponds to the structure of Figure 1 and includes three FMCW radar sensors 10. The transmitted signals 30-1, 30-2, and 30-3 of the three radar sensors 10 are represented by different line thicknesses.

[0071] Within each radar sensor 10, a correspondingly ramp-shaped frequency-modulated transmit signal 30-1, 30-2, or 30-3 is generated. Each transmit signal 30-1, 30-2, and 30-3 contains multiple ramp sequences for each radar sensor 10; in the example shown, there are four ramp sequences. The ramp sequence for the transmit signal 30-1 of the first radar sensor 10 is represented as 1a, 1b, 1c, and 1d in Figure 1. Correspondingly, the ramp sequence for the transmit signal 30-2 of the second radar sensor 10 is represented as 2a, 2b, 2c, and 2d, and the ramp sequence for the transmit signal 30-3 of the third radar sensor 10 is represented as 3a, 3b, 3c, and 3d. Each ramp sequence is represented by a different line type.

[0072] Figure 2 shows only the first two lamps 40 of each lamp sequence. Each shown lamp 40 corresponds to a corresponding single section 42 of the corresponding transmit signal 30-1, 30-2, or 30-3. The frequency-modulated transmit signals 30-1, 30-2, or 30-3 corresponding to the lamps 40 are phase-modulated with harmonic codes for each transmitting antenna element 14 of the corresponding radar sensor and transmitted by the transmitting antenna element 14 as described with reference to Figure 1. The harmonic codes are, for example, identical for all lamp sequences 1a, 1b, 1c, and 1d within one lamp group 44, and the following code setting is selected for each subsequent lamp group 44 (in this regard, each lamp group 44 includes one lamp 40 from each of the lamp sequences 1a, 1b, 1c, and 1d of the radar sensor 10).

[0073] Figure 2 shows an example of radar modulation scheme #1 for three radar sensors 10, each having multiple, for example, four, transmitting antenna elements 14, these transmitting antenna elements 14 are simultaneously active by Doppler division multiplexing (DDM) using harmonic codes. The three sensors 10 operate in interleaved time division multiplexing to keep the time difference between the sensors 10 as small as possible. The ramp-like frequency-modulated transmitting signals 30-1, 30-2, or 30-3 of each of these radar sensors 10 have multiple ramp sequences that are interleaved in time with each other. For example, the transmitting signal 30-1 of the first radar sensor 10 has four ramp sequences 1a, 1b, 1c, and 1d that are interleaved in time with each other. Within each ramp sequence, the ramps are interleaved at a predetermined time interval T r2rThe lamps are staggered and occur in succession. In Figure 2, only the first two lamps of each lamp sequence of these radar sensors 10 are shown. Each measurement cycle includes temporally consecutive intervals 42, each of which is assigned to one of the radar sensors 10 and therefore to the corresponding transmit signals 30-1, 30-2, or 30-3 of the radar sensor 10 in question. Each interval 42 includes one or more lamps 40 of the transmit signal of only the radar sensor 10 assigned to this interval 42. In other words, the first two intervals 42 characterized in Figure 2 include only the lamps 40 of the first radar sensor 10, and these lamps belong to the transmit signal 30-1 of the first radar sensor 10.

[0074] For each of the transmitted signals 30-1, 30-2, and 30-3, the term "lamp group" refers to a group of lamps having the same lamp index in the lamp sequence of the transmitted signal. For example, the transmitted signal 30-1 of the first radar sensor 10 includes a lamp group 44, each containing one lamp 40 from lamp sequences 1a, 1b, 1c, and 1d. Thus, in radar modulation scheme #1, the transmitted signals 30-1, 30-2, and 30-3 of each radar sensor 10 include a sequence 46 of lamp groups 44 arranged sequentially in time, with each lamp group 44 containing one lamp 40 from each of the lamp sequences 1a, 1b, 1c, and 1d of the radar sensor 10.

[0075] Simultaneously, the measurement cycles of the multiple radar sensors 10 are divided into temporally sequential blocks 48. Each block 48 includes one lamp group 44 of each radar sensor 10, that is, one lamp group of each of the transmitted signals 30-1, 30-2, and 30-3 of the radar sensor 10. Block 48 is divided into blocks with a time interval T r2rThese occur in succession with a time delay. In each block 48, the lamps 40 of each lamp group 44 in this block 48 are temporally interleaved with the lamps 40 of each other lamp group 44 in the same block 48. For example, in Figure 2, if we consider a lamp group 44 in the first block 48 that includes lamps 40 with lamp index 1 for the first transmission signal 30-1, that is, lamps with lamp index 1 for lamp sequences 1a, 1b, 1c, and 1d, these lamps 40 are temporally interleaved with the lamps of other lamp groups 44 in this block 48. That is, this block 48 includes a second lamp group 44 that includes lamps with lamp index 1 for lamp sequences 2a, 2b, 2c, and 2d, and a third lamp group 44 that includes lamps with lamp index 1 for lamp sequences 3a, 3b, 3c, and 3d.

[0076] In the example shown, within one block 48, the lamp 40 of the first lamp group 44 has a time difference T relative to the lamp 40 of the second lamp group 44. 12 , and the time difference T of the third lamp group 44 relative to lamp 40 13 It has a time difference T. 12 , T 13 The time interval T of the lamps within each lamp sequence is the time interval T of the lamps within each lamp sequence. r2r It is significantly smaller. There are intervals between some of the lamps in each group of 44 lamps. Further intervals occur between some of the consecutive lamps of different transmission signals 30-1, 30-2, and 30-3.

[0077] The evaluation of the received radar signal corresponds to the flowchart in Figure 3, which is explained below, for determining the relative velocity of radar target 18. In step 60, transmission and reception of radar modulation scheme #1 are performed in accordance with the above explanation. This includes the transmission of phase-modulated transmission signals 30-1, 30-2, and 30-3, and the reception of response signals to the transmitted phase-modulated transmission signals.

[0078] In step 62, a 2D-FFT is performed for each individual sensor on the individual ramp sequences 1a, 1b, 1c, 1d or 2a, 2b, 2c, 2d or 3a, 3b, 3c, 3d. At this time, for each ramp sequence 30-1, 30-2, 30-3 of the transmitted signals of each radar sensor 10, a two-dimensional spectrum is calculated by the two-dimensional Fourier transform, e.g., 2D-FFT, of the baseband signal of the received response signal. This transformation is performed for each ramp 40 in the first dimension and along the ramp index j of the corresponding ramp sequence in the second dimension. Preferably, the amount of each transformation, i.e., the number of bins (sampling points or nodes) for each, is uniform for all spectra for the first and second dimensions, respectively.

[0079] A phase difference occurs between partial measurements based on the relative velocity v of the radar target 18 and the time differences Tab, Tac, and Tad (Figure 2) between partial measurements corresponding to individual ramp sequences. This phase difference between partial measurements is obtained as the phase difference between the complex amplitudes (spectral values) of peaks occurring in the two-dimensional spectrum at the same location. However, based on the relatively large time differences Tab, Tac, and Tad between the corresponding ramps 40 in each ramp sequence, determining the phase difference between partial measurements does not allow for direct inductive reasoning to the relative velocity v. This is because, based on the periodicity of the phase, a single phase difference results in ambiguity in the value of the relative velocity v to which it belongs. In the example, different sensors 10 have the same time differences Tab, Tac, and Tad. In one variation of this example, different time differences Tab, Tac, and Tad are selected for different sensors. In this case, different ambiguities arise for the relative velocity v for each radar sensor, which simplifies the resolution of ambiguity.

[0080] From the obtained two-dimensional complex spectrum, the power spectrum is calculated by squaring the absolute value of each spectral value. These power spectra are then combined point by point to form a unified two-dimensional power spectrum by summing or averaging.

[0081] The peak positions corresponding to radar target 18 within the power spectrum are denoted below as bins k and I, corresponding to the peak positions within the individual spectra. From the first dimension corresponding to bin k of the peak position, a linear relationship between the relative velocity v of the radar target and the interval d is obtained based on the FMCW equation k = 2 / c(dF + f0vT). In the equation, c is the speed of light, F is the ramp displacement, T is the ramp duration of a single ramp, and f0 is the average transmission frequency. If the frequency difference of successive ramps in a sequence is zero, the peak position I in the second dimension contains only information about the relative velocity v of the radar target.

[0082] A linear relationship exists between the relative velocity v and the interval d. In the illustrated example, the information about the relative velocity of the radar target obtained from sampling the Doppler frequency based on a relatively large time interval Tr2r is ambiguous because the Doppler frequency resulting from the relative motion at velocity v is not uniquely sampled due to the relatively large time interval Tr2r of each ramp sequence. In addition to the linear relationship between the relative velocity v and the interval d that arises from the frequency bin k, periodic values ​​of the relative velocity v are obtained from the frequency bin I, thereby providing possible pairs (v,d) of relative velocity and interval for the detected radar target 18.

[0083] Therefore, the ambiguity of the determined velocity v is resolved as explained below. To evaluate the measured phase difference, an ideal measurement control vector a(v) that depends on the relative velocity v is calculated. Measurement vector a mThe vector is appropriately defined, and instead of the expected velocity-dependent complex value, the complex amplitude (spectral value) at the peak position of the calculated two-dimensional spectrum of the partial measurement is used as the component of the vector. Based on this measurement vector and control vector, a normalized likelihood function is defined in the form of the relative velocity spectrum S(v). The maximum of this likelihood function corresponds to the most probable value of the parameter v. This alone may result in an ambiguous relative velocity spectrum S(v). The maximum value of 1 corresponds to the best agreement between the ideal phase shift that occurs for the corresponding relative velocity v and the measured phase shift based on the measurement vector. However, evaluation of the function S(v) is only necessary at points corresponding to the periodic values ​​of relative velocity v obtained from evaluations based on the peak position at bin (k,I). The greatest agreement obtained here with respect to relative velocity v at the point where the function S(v) takes its expected maximum value of 1 corresponds to the actual value of relative velocity v.

[0084] Therefore, the ambiguity arising from the peak position can be resolved by additional information from the phase relationship. Based on the linear relationship, an estimate of the interval d to which the selected estimate of the relative velocity v belongs is determined.

[0085] Time signals (baseband signals) corresponding to different ramp sequences are initially processed separately. Radar target detection is performed within the power spectrum obtained by non-coherent integration. In this case, the ambiguity of velocity v is resolved based on the complex amplitude at the detection and peak positions.

[0086] The above embodiment relates to a conventional radar evaluation without code division multiplexing for different transmitting antenna elements 14-1, 14-2, ... Phase coding of the transmitted signals for each transmitting antenna element by harmonic codes introduces additional ambiguity, which must also be resolved. In the case of code division multiplexing, the transmitters, due to the corresponding code assignments, experience a shift in the peaks in the second dimension of the spectrum. This results in NTX peaks in the spectrum for NTX transmitters. This additional ambiguity must also be resolved.

[0087] Based on the harmonized phase modulation of the transmitted signal, the transmitter shift is equivalent to a bin shift in Doppler dimensions, which is equivalent to the change in the estimated relative velocity v being considered. In a preferred embodiment using harmonic codes for NTX transmitters (NTX transmitting antenna elements per radar sensor 10), this transmitter shift corresponds to an I-bin shift of Δ = N_slow / NTX. Thus, the velocity shift is equivalent to undersampling in Doppler, or in other words, the interval between the evaluated relative velocities is "filled" through the transmitter. In other words, these additional ambiguities based on code division multiplexing must also be estimated when determining the relative velocity of a radar object.

[0088] The method described above can also be extended to perform radar measurements using multiple receiving channels of a single radar sensor. In this case, for each receiving channel, a measurement vector a_m(n) is obtained for the nth channel.

[0089] In Figure 3, step 64 includes step 64-1 for resolving velocity ambiguity and step 64-2 for target detection. Velocity ambiguity resolution may be performed before or after target detection. Velocity ambiguity resolution is carried out as described in U.S. Patent Application Publication No. 2019 / 0353770(A1) or DE102017200317A1 and above with respect to code division multiplexing.

[0090] According to variation 1 of the method in Figure 3, step 66 is performed to correct the time difference between the bistatic and monostatic signals, following step 64. For this purpose, the phase difference between each radar sensor 10, which depends on the relative velocity v of the radar target 18, is taken into consideration.

[0091]

number

[0092] can be determined, for example, corresponding to the following formula.

[0093]

Number

[0094] In the formula, c is the speed of light, f0 is the center frequency of the chirp sequence modulation (lamp), and T 12 is the time shift between the lamp sequences of both sensors (or between the monostatic response and the bistatic response received within one sensor), and v corresponds to the relative speed of the target determined within a single sensor. When the target is far enough away (far field), each of both single sensors measures (almost) the same speed. At close range, the measured speeds of both of these sensors can be different. For the bistatic signal response, the average value of the speeds measured by both (by their respective radar sensors) should be considered.

[0095] Therefore, each phase difference between two of the radar sensors 10 with respect to the radar target 18

[0096]

Number

[0097] is based on the determined estimated value of the relative speed v of this radar target 18, and on the respective time shift T 12 (or T 13 ) between the lamp groups 44 of the respective radar sensors 10 within one block 48.

[0098] The correction of the time shift is, for example, by applying a phase correction to the complex amplitude of the bistatic response signal, for example

[0099]

Number

[0100] This is done by multiplication. Subsequently, the complex amplitudes of the monostatic and bistatic response signals now contain only angular information, and the angle of the radar target can be determined by known methods for angle estimation (e.g., deterministic maximum likelihood estimator, DML). In this case, the complex amplitude is compared with, for example, a measured antenna radiation pattern, and the estimated angle is obtained as the angular position that best matches.

[0101] For angle estimation, step 68 performs digital beamforming for the radar target via the transmit channel assigned to the transmit antenna element 14 and / or the receive channel assigned to the receive antenna element 16 of all radar sensors 10 or a portion of radar sensors 10. Thus, step 68 is a step of coherent MIMO beamforming via the transmit and receive channels of all sensors 10 or a portion of sensors 10.

[0102] According to variation 2 of the method in Figure 3, the implementation of digital beamforming for radar targets is divided into two substeps. In step 68-1, coherent MIMO beamforming is performed for each sensor 10. Thus, in this first substep, digital beamforming for radar targets is performed via the transmit channel assigned to the transmit antenna element and / or the receive channel assigned to the receive antenna element of each radar sensor 10. Here, after step 68-1, in step 69, the output of monostatic results and / or further processing may occur.

[0103] In this case, step 66 is then performed to correct the time difference between the bistatic signal and the monostatic signal. In this case, after step 66, step 68-2 is performed in which coherent coupling of the partial beamforming of a single sensor is carried out. The digital beamforming of each radar sensor 10 is coupled taking into account the phase difference between the radar sensors 10, which depends on the relative velocity v of the radar target 18.

[0104] Figure 4 schematically shows a graph relating the frequencies f of the transmitted signals 30-1, 30-2, and 30-3 of a cooperative radar sensor network 100, which includes three FMCW radar sensors 10, and which can basically correspond to the structure of Figure 1. The transmitted signals 30-1, 30-2, and 30-3 of the three radar sensors 10 are represented here by different line thicknesses.

[0105] Within each radar sensor 10, a correspondingly ramp-shaped frequency-modulated transmit signal 30-1, 30-2, or 30-3 is generated. Each transmit signal 30-1, 30-2, or 30-3 contains multiple ramp sequences for each radar sensor 10; in the example shown, there are four ramp sequences 1a, 1b, 1c, 1d or 2a, 2b, 2c, 2d or 3a, 3b, 3c, 3d. Each ramp sequence is represented by a different line type.

[0106] Figure 4 shows only the first and last ramps 40 of each ramp sequence. Successive ramps 40 belonging to the same transmit signal 30-1, 30-2, 30-3 together constitute a single section 42 of the corresponding transmit signal 30-1, 30-2, or 30-3. The frequency-modulated transmit signals 30-1, 30-2, or 30-3 corresponding to the ramps 40 are phase-modulated with harmonic codes for each transmitting antenna element 14 of the corresponding radar sensor and transmitted by the transmitting antenna element 14.

[0107] Figure 4 shows an example of radar modulation scheme #2 for three radar sensors 10, each having multiple, for example, four, transmitting antenna elements 14, that are simultaneously active by Doppler division multiplexing (DDM) using harmonic codes.

[0108] In this example, instead of time interleaving the transmission operations of the sensors in Figures 2 and 3, the individual sensors perform sequential transmission operations, in which case all sensors receive continuously.

[0109] The ramp-like frequency-modulated transmit signals 30-1, 30-2, or 30-3 of each of these radar sensors 10 have multiple ramp sequences that are interleaved in time with each other. For example, the transmit signal 30-1 of the first radar sensor 10 has four ramp sequences 1a, 1b, 1c, and 1d that are interleaved in time with each other. Within each ramp sequence, the ramps are interleaved in time at predetermined time intervals T r2r The lamps are staggered and occur in succession. In Figure 4, only the first and last lamps of each lamp sequence of these radar sensors 10 are shown. Each measurement cycle includes temporally consecutive intervals 42, each of which is assigned to one of the radar sensors 10 and therefore to the corresponding transmit signals 30-1, 30-2, or 30-3 of the radar sensor 10 in question. Each interval 42 includes all the lamps 40 of the lamp sequence of the transmit signals of only the radar sensor 10 assigned to this interval 42. In other words, the first interval 42 characterized in Figure 4 includes only the lamps 40 of the first radar sensor 10, and these lamps belong to the transmit signal 30-1 of the first radar sensor 10.

[0110] With respect to each of the transmitted signals 30-1, 30-2, and 30-3, the term "lamp group" here again represents a group of lamps having the same lamp index j in the lamp sequence of the transmitted signal. For example, the transmitted signal 30-1 of the first radar sensor 10 includes a lamp group 44, each containing one lamp 40 from lamp sequences 1a, 1b, 1c, and 1d. Thus, in radar modulation scheme #2 as well, the transmitted signals 30-1, 30-2, and 30-3 of each radar sensor 10 include a sequence 46 of lamp groups 44 arranged sequentially in time, with each lamp group 44 containing one lamp 40 from each of the lamp sequences 1a, 1b, 1c, and 1d of the radar sensor 10. There are intervals between some of the lamps in each lamp group 44.

[0111] Simultaneously, the measurement cycles of the multiple radar sensors 10 are divided into temporally consecutive subcycles 70-1, 70-2, and 70-3. Each subcycle 70-1, 70-2, and 70-3 is assigned to one of the radar sensors 10 and includes a group of lamps 44 only for the assigned radar sensor 10. Subcycle 70-1 of the first radar sensor 10 includes the transmission signal of the first radar sensor 10. Correspondingly, subcycles 70-2 and 70-3 each include the transmission signals of the corresponding second radar sensor 10 or third radar sensor 10.

[0112] The subsequent evaluation is carried out according to the method shown in Figure 6, which is explained below. Here, the processing is first performed individually for each subcycle 70, and the differences from the method shown in Figure 3 are explained in particular below.

[0113] In step 80, transmission and reception of radar modulation scheme #2 are performed. Step 80 corresponds to step 60 in Figure 3, the difference being that radar modulation scheme #2, for example, based on Figure 4, is used.

[0114] In this case, for each subcycle 70-1, 70-2, and 70-3, the following processing is performed in the subsequent section 800 of this method. In step 82, a 2D-FFT of the individual lamp sequence is performed for each sensor. Step 82 corresponds in this respect to step 62 in Figure 3. In this example, different sensors 10 or transmit signals 30-1, 30-2, and 30-3 have the same time interval Tr2r and the same lamp duration T_fast for the successive lamps in their respective lamp sequences. In a variation of this example, the time interval Tr2r and / or lamp duration T_fast for different sensors 10 or transmit signals 30-1, 30-2, and 30-3 are selected differently. In this case, different velocity ambiguities arise within these spectra.

[0115] In step 84, velocity ambiguity is resolved either before or after target detection. Step 84 corresponds to step 64 in Figure 3 and may include step 84-1, which resolves velocity ambiguity and corresponds to step 64-1, and step 84-2, which is target detection and corresponds to step 64-2 in any case.

[0116] In the sequential transmission operation of individual sensors 10, the accuracy of speed measurement for a single sensor is generally insufficient to enable coherent processing of the transmitted signals of all sensors. Instead, separate evaluations of so-called "partial cycles" are performed, in which case the transmitted signal of the active sensor and the received signals (RX) of all sensors can be evaluated.

[0117] In variation 1 of section 800, step 86 performs coherent MIMO beamforming through all transmit and receive channels of the current subcycle. Thus, digital beamforming is performed for the radar target 18 and for the applicable subcycle 70 through the transmit channel assigned to the transmit antenna element 14 and / or the receive channel assigned to the receive antenna element 16 of the applicable radar sensor 10. Step 86 also involves, among other things, determining the angle estimate of the radar target 18, and this determination of the angle estimate of the radar target includes performing digital beamforming for the radar target.

[0118] In variation 2 of section 800, the implementation of digital beamforming for a partial cycle is divided into a partial step 86-1 of per-sensor coherent MIMO beamforming and a partial step 86-2 of coherent coupling of single-sensor partial beamforming. Following step 86-1, a monostatic result output and / or further processing may occur in step 87. The coupling of partial beamforming in partial step 86-2 can also be done taking into account the time difference between the signals of these sensors 10, i.e., it may include time difference correction.

[0119] A second section of this method may follow this first section 800. According to one variation of this method, in step 90, the result is output for each subcycle. The sequential transmission operation of each sensor 10 processes each "subcycle," resulting in an update frequency of the detected target that increases with the number of sensors.

[0120] Simultaneously, in these partial cycles 70, the activation of the transmitter (TX), i.e., the transmitting antenna element 14, varies, resulting in various virtual antenna arrays with different ambiguous characteristics during angle estimation. This is schematically illustrated in Figure 5.

[0121] Figure 5 shows, at the top, the configuration of the transmitting antenna element array TX of the first radar sensor 10 that is actively transmitting and the three receiving antenna element arrays RX of the three radar sensors 10 that are receiving, corresponding to partial cycle 70-1. A corresponding virtual antenna array is generated.

[0122] Figure 5, at the bottom, shows the configuration of the transmitting antenna element array TX of the second radar sensor 10 that is actively transmitting and the three receiving antenna element arrays RX of the three radar sensors 10 that are receiving, corresponding to partial cycle 70-2. A corresponding separate virtual antenna array is generated.

[0123] These sensors 10 may have different array configurations TX and / or RX, for example, array configurations with different spacing between antenna elements. According to one alternative variation, or in addition to the processing in step 90, this method may include steps 92 and 94 as described below.

[0124] In step 92, non-coherent averaging of the subcycles 70 is performed. This non-coherent averaging of the subcycles can be performed on the spectral plane before target detection, or by averaging the angular spectra after target detection. In the first case, improved detection performance is achieved by power-averaging the range-Doppler spectra. However, the complex amplitude of the unaveraged spectra is then again required so that individual angular estimates / angular spectra can be determined for each subcycle. In the second case, detection is first performed within a single spectrum, and then angular spectra are calculated for each subcycle 70 (e.g., using a DML estimator). In this case, the angular spectra of the subcycles can then be non-coherently averaged.

[0125] Therefore, this method includes, in step 92, the combination of digital beamforming of the radar sensor by non-coherent averaging of the digital beamforming of the radar sensor and the applicable partial cycle, or by averaging the angular spectra of the radar target obtained by digital beamforming of the radar sensor and the applicable partial cycle. In step 94, the output of the combined result is performed. Therefore, non-coherent averaging may be performed already on the spectral plane before target detection, or by averaging the angular spectra after target detection.

[0126] In the examples in Figures 2 and 4, the ramp sequences of individual transmitted signals 30-1, 30-2, and 30-3 each have the same parameters, e.g., ramp duration T_fast, ramp deviation F_fast, and ramp center frequency f0, and therefore the same frequency transition. Figure 2 exemplifies the following parameters: ramp duration T_fast, ramp deviation F_fast, and ramp center frequency f0. The ramp center frequency here corresponds to the average transmitted frequency f0. In another example, the ramp center frequency of a ramp may rise or fall during the transmission of a ramp sequence. For example, successive ramps within a single ramp sequence may have the same difference in ramp center frequency. Given the time T_slow (corresponding to the duration of the measurement cycle) of a single ramp sequence and the frequency deviation F_slow of the ramp center frequency within this entire ramp sequence, the ramp slope s_slow of the ramp center frequency can be calculated by s_slow = F_slow / T_slow.

[0127] In particular, ramps 40 having the same ramp index j within interleaved ramp sequences may each have the same ramp gradient F_fast / T_fast.

Claims

1. A method for determining the relative velocity (v) of a radar target (18), The step involves transmitting a phase-modulated transmission signal (30-1; 30-2; 30-3) using multiple transmitting antenna elements (14) of each radar sensor (10), wherein each radar sensor generates a frequency-modulated transmission signal (30-1; 30-2; 30-3) in a ramp shape, and the transmission signal (30-1; 30-2; 30-3) has a sequence (1a; 1b; 1c; 1d; ...) of multiple time-interleaved ramps (40), and the ramps (40) within each of the sequences (1a; 1b; 1c; 1d; ...) are at predetermined time intervals (Tr2 Step r) is performed sequentially with a time stagger, and the phase of the generated ramp-shaped frequency modulated transmit signals (30-1; 30-2; 30-3) is phase modulated by a code for each transmitting antenna element (14) of the radar sensor (10), and the measurement cycles of the plurality of radar sensors (10) include time-sequential intervals (42), each of which is assigned to one of the radar sensors (10), and includes one ramp (40) or more ramps (40) of the transmit signals (30-1; 30-2; 30-3) only for the radar sensor (10) assigned to the interval, The steps include receiving a response signal to the transmitted phase-modulated transmission signal (30-1; 30-2; 30-3), The steps include: calculating a two-dimensional spectrum for each sequence (1a; 1b; 1c; 1d; ...) of the transmitted signal (30-1; 30-2; 30-3) of each radar sensor (10) by performing a two-dimensional Fourier transform of the baseband signal of the received response signal, wherein the transformation is performed per lamp in the first dimension and along the lamp index of the sequence for the lamp in the second dimension; The steps include determining the value of the periodic relative velocity (v) of the radar target (18) at a predetermined velocity period, based on the peak in one of the calculated two-dimensional spectra, The steps include determining an estimated value of the relative velocity (v) of the radar target (18) based on the agreement between the phase relationships between two-dimensional spectral values ​​at the same position and the phase relationships that can be expected for a plurality of the determined relative velocity (v) values, A method of having.

2. Each of the aforementioned radar sensors (10) transmits a sequence (46) of lamp groups (44) arranged in chronological order, and each lamp group (44) includes one lamp (40) from each of the lamp sequences (1a; 1b; 1c; 1d; ...) of the transmit signals of the radar sensors (10). The method according to claim 1, wherein the measurement cycle of the plurality of radar sensors (10) includes a block (48) that is sequential in time, and each block (48) includes a group of lamps (44) from each of the radar sensors (10).

3. The method according to claim 2, wherein within each block (48), the lamps (40) of one lamp group (44) of the block (48) are temporally interleaved with the lamps (40) of each other lamp group (44) of the block (48).

4. The method according to any one of claims 1 to 3, wherein the transmission signals (30-1; 30-2; 30-3) of the plurality of radar sensors (10) have different time differences (Tab; Tac; Tad) between the sequences (1a; 1b; 1c; 1d; ...) of the lamps (40).

5. The process includes the step of performing digital beamforming for a radar target (18) via the transmit channels assigned to the transmit antenna elements (14) and / or the receive channels assigned to the receive antenna elements (16) of at least two of the plurality of radar sensors (10), wherein, when performing the digital beamforming, the phase difference between the radar sensors (10) is taken into consideration, depending on the relative velocity (v) of the radar target (18), and for the radar target (18), the phase difference between each of the two of the radar sensors (10) is based on at least one estimate of the relative velocity (v) of the radar target (18), and the time difference (T) between the ramp groups (44) of the two radar sensors (10) (within one block). １２ The method according to any one of claims 1 to 4, determined based on ;. . .).

6. The method of claim 5, comprising the step of determining the angle estimation of the radar target (18), wherein determining the angle estimation of the radar target (18) includes performing the digital beamforming for the radar target (18).

7. The method according to claim 5 or 6, wherein the implementation of the digital beamforming for the radar target (18) comprises a first partial step (68-1) wherein the digital beamforming for the radar target (18) is performed via a transmit channel assigned to the transmit antenna element (14) and / or a receive channel assigned to the receive antenna element (16) of each of the at least two radar sensors of the plurality of radar sensors (10), and a second partial step (68-2) wherein the digital beamforming of each of the at least two radar sensors (10) is coupled, taking into account the phase difference between the radar sensors (10) that depends on the relative velocity (v) of the radar target (18).

8. Each of the aforementioned radar sensors (10) transmits a sequence (46) of lamp groups (44) arranged in chronological order, and each lamp group (44) includes one lamp (40) from each of the lamp sequences (1a; 1b; 1c; 1d; ...) of the transmit signals of the radar sensors (10). The method according to claim 1, wherein the measurement cycle of the plurality of radar sensors (10) includes temporally consecutive partial cycles (70-1; 70-2; 70-3), each partial cycle (70-1; 70-2; 70-3) is assigned to one of the radar sensors (10), and the lamp group (44) includes only the assigned radar sensor (10).

9. The method according to claim 8, wherein the transmission signals (30-1; 30-2; 30-3) of the plurality of radar sensors (10) have different time intervals (Tr2r) in which the lamps (40) in each of the sequences (1a; 1b; 1c; 1d; ...) of the lamps (40) are staggered and successive, and / or have different lamp durations (T_fast).

10. The method according to claim 8 or 9, further comprising the step of performing digital beamforming for a radar target (18) and for applicable partial cycles (70-1; 70-2; 70-3) via a transmit channel assigned to the transmit antenna element (14) and / or a receive channel assigned to the receive antenna element (16) of each of the plurality of radar sensors (10).

11. The method according to claim 10, comprising the step of determining an angle estimate of the radar target (18), wherein determining the angle estimate of the radar target (18) includes performing the digital beamforming for the radar target (18).

12. The digital beamforming is performed for the radar target (18) and for each applicable partial cycle (70-1; 70-2; 70-3) via the transmit channel assigned to the transmit antenna element (14) and / or the receive channel assigned to the receive antenna element (16) of at least two radar sensors of the plurality of radar sensors (10), The method according to claim 10 or 11, further comprising the step (92) of combining the digital beamforming of the at least two radar sensors (10) by non-coherent averaging of the digital beamforming of the at least two radar sensors and the applicable partial cycles (70-1; 70-2; 70-3), or by averaging the angular spectra of the radar target (18) obtained by the digital beamforming of the at least two radar sensors (10) and the applicable partial cycles (70-1; 70-2; 70-3).

13. The method according to any one of claims 1 to 12, wherein the reception of response signals to the transmitted phase-modulated transmission signals (30-1; 30-2; 30-3) includes, namely, for each phase-modulated transmission signal (30-1; 30-2; 30-3) transmitted by the transmitting antenna element (14) of the radar sensor (10), the response signal is received by at least one receiving antenna element (16) of the same radar sensor (10), and the response signal is received by at least one receiving antenna element (16) of another radar sensor among the plurality of radar sensors (10).

14. A radar sensor network (100) for determining the relative velocity of a radar target, comprising a plurality of radar sensors (10) and at least one evaluation unit (20, 12), Each of the plurality of radar sensors (10) Multiple transmitting antenna elements (14), A signal generator (17) is designed to generate a ramp-shaped frequency-modulated transmission signal (30-1; 30-2; 30-3) and provide it to the transmitting antenna elements, wherein the transmission signal (30-1; 30-2; 30-3) has a sequence (1a; 1b; 1c; 1d; ...) of multiple time-interleaved ramps (40), where the ramps (40) in each sequence (1a; 1b; 1c; 1d; ...) are time-shifted by a predetermined time interval (Tr2r), and the phase of the generated ramp-shaped frequency-modulated transmission signal (30-1; 30-2; 30-3) is phase-modulated by a code for each transmitting antenna element (14) of the radar sensor (10). At least one receiving antenna element (16) designed to receive a response signal to the transmitted phase-modulated transmission signal (30-1; 30-2; 30-3), Includes, A radar sensor network (100) is designed in which the evaluation units (20, 12) are configured to determine a value of the relative velocity (v) of a radar target (18) that is periodic at a predetermined velocity period, based on a peak in one of the two-dimensional spectra calculated for each sequence of the transmitted signals (30-1; 30-2; 30-3) of each radar sensor (10), and to determine an estimate of the relative velocity (v) of the radar target (18) based on the agreement of the phase relationships between the two-dimensional spectral values ​​at the same location with the phase relationships that can be expected for a plurality of the determined relative velocity (v) values.

15. The radar sensor network according to claim 14, wherein each of the plurality of radar sensors (10) includes a plurality of receiving antenna elements (16).

Images