Method for operating Doppler multiplexing method in radar network and radar network

Abstract

The invention relates to a method for operating a Doppler multiplexing method (DDM) in a multiple-input multiple-output (MIMO) radar network, in which the signals of at least a first radar (Rad1) and a second radar (Rad2) in the MIMO radar network, which are explicitly encoded by means of the Doppler multiplexing method (DDM), are convolved using corresponding DDM codes, and the signals which can be explicitly assigned to the radars (Rad1, Rad2) are determined. Rad2) of the signal (Rad1). Subsequently, the occurrence of the explicitly assigned signal in the monostatic path, in the bistatic path and in all paths is determined by means of a search algorithm, so that a robust MIMO angle estimation can be made with respect to any aperture. The invention also relates to a radar network.

Description

Field of the Invention

[0001] The present invention relates to a method for angle estimation based on signals of a multiple-input multiple-output radar network (MIMO) operating in a Doppler multiplexing method (Doppler-Division-Multiplexing, DMM).

[0002] The present invention particularly relates to a method for operating a Doppler multiplexing method in a radar network, which is used in a driver assistance system of a motor vehicle for environmental detection. Background Art

[0003] Modern radar systems are particularly constructed as multiple-input multiple-output (MIMO) radar networks for driver assistance systems of motor vehicles and include a plurality of transmitters (Tx) and receivers (Rx). In order to split and distinguish a higher number of transmitters on the receiving side, such a radar network can be operated in a Doppler multiplexing method (DDM method, Doppler-Division-Multiplexing). Thus, an extension of the minimum measurement time required for splitting the transmitters on the receiving side can be prevented, as such an extension may occur, for example, in a radar network operated using a time division multiplexing method (TDM method, Time Division Multiple Access). DE 10 2020202 500 A1 describes a TDM method for shortening the required minimum measurement time. With the DDM method, the transmit power and thus also the energy emitted simultaneously can be increased in proportion to the measurement time. This results in an improved signal-to-noise ratio during data processing, which in turn enables an extension of the effective range.

[0004] Applying the Doppler multiplexing method in a radar network presents a particular challenge. In this case, it must be taken into account that due to the spatial separation between the individual radar devices, different radial velocities and ranges may be measured for the same target. This results in distortion of the codes of the Doppler multiplexing method. Thereby, the decryption and evaluation of the data are impaired, especially in terms of angle estimation when using multiple input and output channels (MIMO).

[0005] To improve the discrimination of the transmitters, the corresponding transmitted signals can be pre-coded for each transmitter by means of a phase shift, so that the received signals can be assigned to the transmitters by their phase shifts. Such a method is described, for example, in US2022 / 0171049 Al. Summary of the Invention

[0006] The object of the present invention is to provide a method that allows for a robust decoding of the codes of the Doppler multiplexing method within a radar network and thus enables an accurate angle estimation.

[0007] This object is solved according to the invention by the method given.

[0008] The method according to the invention for operating a Doppler multiplexing method (DDM) in a multiple-input multiple-output (MIMO) radar network comprises the following steps:

[0009] Generating and transmitting an encoded signal to at least a first and a second radar in the MIMO radar network by means of the Doppler multiplexing method, wherein each radar comprises at least one transmitter and one receiver, and wherein the encoded signal has a corresponding DDM code that clearly identifies each radar;

[0010] Receiving and preprocessing the reflected signal for digital processing and conversion into a range-Doppler-velocity matrix;

[0011] Performing a threshold detection for each range-Doppler matrix to determine the cells having potential signals of the corresponding radar, wherein the threshold detection performs a binary transformation on these signals;

[0012] Convolving each of the explicit DDM codes in the explicit DDM code with the corresponding potential binary signal and unambiguously assigning the corresponding signal to one of the radars;

[0013] Determining the signals whose signal parameters overlap for each radar in the bistatic path and unambiguously assigned signals for each radar in the monostatic path, wherein the signals of the corresponding monostatic path are different from the signals deviating from the bistatic path;

[0014] Determining the signals that are unambiguously assigned both in the bistatic path and in each monostatic path, wherein the signals of the monostatic path are determined starting from the corresponding signals of the bistatic path, and

[0015] Performing MIMO angle estimation using the full aperture of the virtual MIMO array formed by the signals that are unambiguously assigned in all paths.

[0016] According to a preferred embodiment, the method further comprises the step of "determining the signals that are assigned only in the bistatic path"; the step of "performing MIMO angle estimation using the bistatic aperture of the first and second radars", and / or the method further comprises the step of "determining the signals that are assigned only in one of the monostatic paths"; the step of "performing MIMO angle estimation using at least one of the monostatic apertures of the first radar and the monostatic aperture of the second radar".

[0017] The radar network may include more than two radars, wherein each DDM code uniquely identifies the corresponding radar.

[0018] Each radar may include a plurality of transmitters, wherein the DDM code includes a distinct phase shift of the signals of each transmitter of the corresponding radar. The distinct phase shift causes the measured signals to be assigned to the corresponding transmitters in such a way that no transmitter delivers the same Doppler-velocity-bin of the range-Doppler velocity-matrix for a target with the same velocity. Here, the range-Doppler velocity-matrix sorts the received signals into specific bins according to their velocity and their range. Thus, the velocity of a target can be uniquely reconstructed without a common evaluation of the DDM codes of all radars of the radar network. For example, the phase shifts may be non-overlapping, and additionally for example the phase shifts may be relatively prime and include relatively prime differences of the phase shifts between the transmitters of the corresponding radar or any combination thereof. In this case, the DDM codes of the individual radars are distinguished from each other in such a way that the DDM code uniquely identifies each radar, that is, no DDM code is the same as another DDM code. When using radars each having only one transmitter, the DDM code corresponds to the distinct phase shift of this transmitter of the corresponding radar.

[0019] By means of a search algorithm, starting from the corresponding signals of the bistatic path, the determination of the signals that have been uniquely assigned both in the bistatic path and in each monostatic path is performed.

[0020] In this case, the signals that have been uniquely assigned both in the bistatic path and in each monostatic path can be determined starting from the corresponding signals of the bistatic path by means of a point-symmetric search.

[0021] In this case, the search range of the point-symmetric search can be expanded by one unit for each dimension from zero bin up to the maximum possible deviation.

[0022] The method may be implemented using linearly frequency-modulated chirp signals or exponentially frequency-modulated chirp signals or piecewise frequency-modulated chirp signals as sequential transmit signals.

[0023] In this case, the maximum possible deviation can be adapted so that the deviation is adapted to the distance-based frequency in the sequential frequency modulation of the piecewise frequency modulation.

[0024] Accordingly, a preferred expansion of the present invention includes the implementation of the method according to the time division multiplexing (TDM) mode. In this method, the corresponding DDM codes of the first radar and the second radar are exchanged in the first and second time windows of the time division multiplexing mode. The method is implemented separately for the first time window and the second time window, and the results of the corresponding time windows are compared with each other, and only the signals that are the same for each time window are considered. The same signals are those whose parameters are the same for the corresponding cells of the range-Doppler velocity matrix. This embodiment assumes that the first and second radars transmit with the same number of transmitters.

[0025] Preferably additionally, the results of the first and second time windows are compared with each other before the step of performing MIMO angle estimation, and the MIMO angle estimation is only performed for the signals taken into account.

[0026] The time distance between the time windows is only a few microseconds long. Therefore, the signal parameters in the two time windows are approximately the same and can be output as the same cells by transformation into the range-Doppler velocity matrix. Accordingly, two adjacent time windows deliver the same results for the correctly assigned signals.

[0027] This expansion has the advantage that false-positive measurements, in which the DDM codes are accidentally formed by the echoes of the reflections of the signals of the individual radars, but these echoes come from various measured targets, do not occur very often. Thereby, a false signal that is not based on a single measured target is constructed. By the method steps of "determining the signals clearly assigned in the bistatic path (BP)" and "determining the signals that are clearly assigned both in the bistatic path (BP) and in each monostatic path (MP 1 , MP 2 ), false-positive measurements cannot always be avoided, or rather, false-positive measurements cannot always be detected by these method steps. In particular, in the step of "determining the signals that are assigned only in one of the monostatic paths (MP1, MP2)", it is not always clear whether the signal is a signal constructed only in the monostatic path or whether the signal is an artifact of multiple signals based on various measured targets. By this preferred expansion, false-positive detection is largely prevented because false signals accidentally formed by the echoes of the reflections of various measured targets must be constructed for each DDM code in the DDM code in order to be taken into account.

[0028] When implementing such an embodiment according to the invention in a radar network with more than two radars, the DDM codes of the individual radars can be pairwise exchanged for two time windows each. This can be carried out continuously (gleichbleibend) for all time windows, or the DDM codes can be exchanged for two time windows each for two previously determined or randomly selected radars. The DDM codes of a large number of radars can also be shifted by one radar each for each time window, where this shift must be taken into account during evaluation.

[0029] In this case, the MIMO angle estimation of the data under consideration is advantageously carried out for each time window using the correction of the phase difference based on speed. This can be achieved without problems by known methods.

[0030] The invention also includes a radar network having at least a first and a second radar. The method according to a preferred embodiment or expansion variant is implemented using this radar network. Thus, each radar includes at least one transmitter and receiver, which can be implemented as a conventional radar antenna, and additionally they can also be implemented such that each transmitter is also a receiver at the same time.

[0031] In this case, the implemented features can obviously be combined with each other arbitrarily if this is technically possible. Description of the Drawings

[0032] In the following, the preferred embodiments of the invention are explained in more detail based on the drawings. The drawings show:

[0033] Figure 1 Schematic diagram of a radar system operating using the method according to the invention;

[0034] Figure 2 Range-Doppler velocity-matrix with an example of DDM codes;

[0035] Figure 3 Overview of the steps of the first embodiment of the method;

[0036] Figure 4 Evaluation of the measurement results obtained according to the first embodiment of the method;

[0037] Figure 5 Example of signals assigned in all paths;

[0038] Figure 6 Time pattern of the time multiplexing method according to a preferred expansion variant of the method according to the invention;

[0039] Figure 7 Execution of the method according to the preferred expansion variant. Detailed Description of the Invention

[0040] Figure 1 Schematic diagram showing a radar system 10 operating using the method according to the invention. Here, the radar system 10 is arranged, for example, on a vehicle F and includes a first radar Rad 1 and a second radar Rad 2 . Here, each radar Rad 1 、Rad 2 includes at least one transmitter (Tx, not shown) and a receiver (Rx, not shown). Here, the transmitter can simultaneously be a receiver, or the transmitter and the receiver are implemented as separate antennas. Preferably, each radar includes a plurality of transmitters and receivers.

[0041] Here, the radars Rad 1 and Rad 2 are arranged at a distance d from each other with respect to their reference points. In Figure 1 , the radar system 10 detects a point target P moving at a relative speed v. The radars Rad 1 and Rad 2 from each other results in different measured radial velocities v 1 、v 2 and different distances R 1 、R 2 to the point target P.

[0042] By combining each transmitter with each receiver, a virtual array can be generated that consists of virtual channels and is based on the sum of the propagation time differences of the signals from the transmitter to the point target and from the point target to the receiver. This makes it possible to increase the aperture of the radar system 10 and thus makes it possible to increase the resolution of the radar system. In this case, the received signals must be clearly assigned to the transmitters, and the DDM method can be used for said assignment.

[0043] It is assumed for this method that: the speed v and the distance R are approximately the same for all virtual channels of the radar. Thus, in the case where radar 1 (Rad 1 ) has a distance R 1 and a speed v 1 and radar 2 (Rad 2 ) has a distance R 2 and a speed v 2 the following applies:

[0044] v 1 =v 2 and R 1 =R 2 .

[0045] In the bistatic path (BP) to be measured, i.e., in the case of receiving a signal using a receiver that is not simultaneously a transmitter, the speed is obtained:

[0046] and the distance

[0047] Here it is assumed that v 1 ≥ v mix ≥ v 2 and R 1 ≥ R mix ≥ R 2 .

[0048] In this case it is clear to the person skilled in the art that the distance and speed of the monostatic path (MP 1 , MP 2 ) (where the transmitter is simultaneously a receiver) of the radar are different from those of the bistatic path, either upwards or downwards. Thus, it applies:

[0049] and

[0050] For ΔR, ΔR max ∈ [0; d / 2] applies as the upper boundary value of this deviation, where d is the distance between the reference points of the two radars. For Δv, Δv max ∈ [0; v lim / 2] applies for most detected targets, where there is no theoretical upper boundary value. Thus, v lim is determined as the boundary value and corresponds to the boundary value below which most of the measured speeds lie.

[0051] The corresponding DDM codes (DC 1 , DC 2 ) of each radar (Rad 1 , Rad 2 ) with multiple transmitters can include a distinct phase shift of the signal of each transmitter (Tx) of the corresponding radar (Rad 1 , Rad 2 ) for unambiguous assignment. The DDM code can in particular include the relatively prime differences of the phase shifts between the transmitters of the corresponding radar. For example, the two differences of the phase shifts of two transmitters of one radar can be relatively prime.

[0052] Figure 2 a shows the range-Doppler velocity-matrix a) presented as an example of such a DDM code. The radar identified by the DDM code includes two transmitters Tx 1 and Tx 2 . Transmitters Tx 1 and Tx 2The signals are encoded in such a way that they are phase - shifted disjointly. From Tx 1 to Tx 2 , the difference in phase shift between two consecutive signals reaches 5 / 8 of the maximum speed v max . From Tx 2 to Tx 1 , considering periodic extension, the difference in phase shift between two consecutive signals reaches 3 / 8 of the maximum speed v max . Thus, the differences in phase shift are relatively prime to each other and the DDM code is unambiguous. Due to the periodic extension or the circular array structure in the Doppler dimension, a unique velocity is obtained in each measurement and subsequent convolution with the DDM code, and thus the measured velocity can be uniquely assigned to the transmitter (Tx).

[0053] Figure 2 b shows such a measurement of threshold detection according to which the threshold detection enables a binary transformation and enables display as a range - Doppler velocity - matrix. In the range - Doppler velocity - matrix b), a unique velocity v Figure 2 is obtained by convolution using the DDM code in est a. This velocity also enables the radar transmitters Tx 1 , Tx 2 to be assigned to the corresponding cells. In other words, the unique velocity of the target is 3 / 8v max , which is given by the convolution using the DDM code and gives the measured velocity 3 / 8v 1 for Tx max and the measured velocity 3 / 8v 2 for Tx max .

[0054] In Figure 2 c, a contrary example is shown. The DDM codes of the two transmitters Tx 1 and Tx 2 are not unambiguous because the difference in phase shift between the transmitters is not relatively prime as long as the measurement delivers ambiguous results as shown in Figure 2 d. Therefore, the velocity v est cannot be uniquely determined, and without comparing the overall measured velocity of the radar, it is impossible to uniquely assign the transmitters Tx 1 and Tx 2 to the measured cells. This slows down the evaluation of the measurement and shows the drawback solved by the method according to the invention.

[0055] Figure 3Shows an overview of the first step before the evaluation step regarding the first embodiment of the method. In step S1, for each radar Rad of the radar network operating using the method 1 、Rad 2 Generate a clearly distinguishable and encoded signal corresponding to the radar Rad 1 、Rad 2 . This can be achieved, for example, by a distinct phase shift of the signals of the respective transmitters of each radar as described. In step S2, each radar transmits the signal specifically encoded for the corresponding radar Rad 1 、Rad 2 and receives all the reflected signals within its angular range. Subsequently, these signals are processed for digital processing and converted into a range-Doppler velocity matrix (step S3).

[0056] By means of the threshold detection performed in step S4, the detected signals are binary-transformed and values 0 or 1 are obtained. For example, in this case, the threshold of the amplitude is used in the frequency representing the Doppler velocity. Thereby, potential signals can be filtered out and signals with too small an amplitude can be excluded. Subsequently, such binary-transformed and filtered range-Doppler velocity matrices are evaluated (step S5).

[0057] Figure 4 Shows such an evaluation. In step S5.1, the range-Doppler velocity matrices (range-Doppler-matrix, RDM) corresponding to the radar Rad Figure 4 、Rad 1 、Rad 2 generated as shown in 1 、DC 2 are obtained. In step S5.2, each of the two radars' distinct DDM codes DC 1 and DDM code DC 2 are used to convolve this binary range-Doppler velocity matrix, that is, here the DDM code DC 1 and DDM code DC 2 are used to convolve the range-Doppler velocity matrix of Rad 1 , and similarly, the DDM code DC 1 and DDM code DC 2 are used to convolve the range-Doppler velocity matrix of Rad 2 . As shown in Figure 2 , for each signal, exactly one velocity v est and a distinct assignment of the radar that transmitted the measured signal as the basis of the encoded signal are obtained through the distinct DDM code. Each potential signal among the potential signals in each range-Doppler velocity matrix is accordingly assigned to a velocity vest and a radar.

[0058] Subsequently, in step S5.3, the following clearly assigned signals are determined: the signal is measured in the bistatic path (BP), and the signal parameters (R est and v est ) for each radar overlap (R mix and v mix ), that is to say, the signal is the same for each radar Rad 1 、Rad 2 . This means that for each radar, the signal parameters (R est and v est ) are in the same cell in the range-Doppler velocity matrix, and it can be said that the signal parameters point to the same place. This improves the robustness of the method.

[0059] Also in step S5.4, the clearly assigned signals measured in the monostatic path are determined. For example, using the DDM code DC 1 of Rad 1 to perform convolution on the range-Doppler velocity matrix of radar Rad 1 gives the signal constructed in the monostatic path MP 1 of Rad 1 .

[0060] In order to enable MIMO angle estimation for the complete virtual aperture, in step S5.5, the signals that are clearly assigned both in the bistatic path (BP) and in each monostatic path (MP 1 , MP 2 ) are determined. This determination is implemented by means of a search algorithm. The search algorithm starts from the corresponding signals (R mix and v mix ) in the bistatic path to search for the signals (R x , v x ) in the monostatic path. As described, the deviation of the signals for the corresponding monostatic paths applies and wherein the boundaries of this deviation are located as ΔR max ∈ [0;d / 2] and Δv max ∈ [0;v lim / 2] as the upper boundary values. The search algorithm starts from the point R est , v est in the range-Doppler velocity matrix and searches for the signals (R x , v x)。In this case, the search range, for example for each dimension, extends from the zero cell up to the maximum possible deviation (ΔR max , Δv max ) by one cell each.

[0061] Figure 5 Such a signal is shown. The bistatic signal (R mix , v mix ) is constructed in the bistatic path BP, and the monostatic signals (R 1 , v 1 ; R 2 , v 2 ) are constructed symmetrically about this bistatic signal at the midpoints of their range-Doppler velocity-matrices, respectively. They have a deviation of one range cell (ΔR) and two velocity cells (Δv). The search algorithm first searches, for example, in the cells directly surrounding the bistatic signal for the monostatic signals, and subsequently widens its search range by one cell each. In this way, the search algorithm searches for the monostatic signals (R 1 , Rad 2 of the radars Rad 1 , v 1 ; R 2 , v 2 ) and identifies the point symmetry.

[0062] In the current case, the deviation is small, but the search algorithm also takes note here of the periodic extension or the circular array structure in the Doppler dimension in order to correctly assign the monostatic signals to the bistatic signals.

[0063] The determination of the signals assigned only in the monostatic path in step S5.4 can alternatively also be carried out via step S5.5, in which all monostatic signals that are not assigned in all paths or not assigned in the bistatic path are determined.

[0064] Furthermore, in step S5.7, the signals assigned only in the bistatic path can be determined.

[0065] By completely assigning the signals to the corresponding radars or the corresponding transmitters and receivers, the signals assigned in all paths can subsequently be used to perform MIMO angle estimation (step S5.8) with respect to the complete virtual aperture. This results in improved angle resolution and thus in more precise results. In this case, the monostatic aperture of the first radar Rad 1 , the monostatic aperture of the second radar Rad 2 and the monostatic apertures of the first and second radars Rad 1 , Rad 2Perform MIMO angle estimation for each of the bistatic apertures as required. In this case, the Doppler velocity is output within the full unambiguous range.

[0066] A preferred extension of the present invention includes the allocation of the method in a Time Division Multiplex (TDM) mode.

[0067] Figure 6 Illustrates this preferred extension of the method, which has an allocation in a Time Division Multiplex (TDM) mode. The TDM mode has a plurality of time windows Z, which are currently divided into a first time window Z that is executed alternately 1 and a second time window Z 2 . It is obvious here that the present invention is not limited to two time windows, but may include any number of time windows that can be technically implemented. Each radar Rad 1 、Rad 2 sends signals encoded with a specific DDM code in each time window, as previously described. Here, the DDM codes DC 1 、Rad 2 used for Rad 1 、DC 2 are swapped for each of the two time windows Z 1 ,Z 2 . In the first time window Z 1 , the signal of Rad 1 is encoded with the DDM code DC 1 , and the signal of Rad 2 is encoded with the DDM code DC 2 . In the second time window, the signal of Rad 1 is encoded with the DDM code DC 2 , and the signal of Rad 2 is encoded with the DDM code DC 1 . The time windows Z 1 、Z 2 are alternately implemented during the execution of the time division multiplexing method. Therefore, this swap is repeated for each time window.

[0068] Figure 7 Illustrates the execution of the preferred extension of the method. The steps described for Figure 3 and Figure 4 and not individually shown here again also apply to this preferred extension. The difference lies in steps S1 and S6. In step S1, for Rad 1 and Rad 2The signals use their corresponding DDM codes DC 1 , DC 2 for encoding, where this encoding process is, for example, Figure 6 as described for the time window Z 1 , Z 2 exchanged for each time window. In step S6, the assigned signals from each of steps S5.3 to S5.7 are compared according to their parameters (R x , v x ) and only those signals that are the same for each time window in the time window are selected. Subsequently, only these selected data are used to execute step S5.8, in this way classifying false positive measurements, because the probability that these false positive measurements are the same in two time windows is extremely small as described. Therefore, the robustness of the method can be further improved. In this case, for example, it may also be possible that step S5.8 is carried out before step S6 and only those values are further processed from the MIMO angle estimation of the two time windows: these values correspond to those signals that are selected in step S6.

[0069] Although the clear range of the speed of the method, which is first used for clear speed determination in the second embodiment, is reduced by a factor of two by the alternately implemented time windows, this disadvantage can be compensated for in a known manner by correspondingly implementing the TDM method.

[0070] List of reference signs and parameters

[0071] 10 Radar system

[0072] F Vehicle

[0073] d Rad 1 and Rad 2 distance to

[0074] P Point target

[0075] Rad 1 Radar 1

[0076] Rad 2 Radar 2

[0077] Tx 1 Transmitter

[0078] Tx 2 Transmitter

[0079] DC 1 DDM code 1

[0080] DC 2 DDM code 2

[0081] BP bistatic path

[0082] MP 1 Rad 1 monostatic path of

[0083] MP 2 Rad 2 monostatic path of

[0084] v radial velocity

[0085] v 1 by Rad 1 measured radial velocity

[0086] v 2 by Rad 2 measured radial velocity

[0087] R 1 radar Rad 1 distance to P

[0088] R 2 radar Rad 2 distance to P

[0089] v mix radial velocity in the bistatic path

[0090] R mix distance in the bistatic path

[0091] v est definite radial velocity

[0092] ΔR deviation of distance

[0093] Δv deviation of velocity

[0094] ΔR max maximum deviation of distance

[0095] Δv max maximum deviation of velocity

[0096] TDM time-division multiplexing method

[0097] Z 1 TDM time window 1

[0098] Z 2 TDM time window 2

Claims

1. A method for operating a Doppler multiplexing method (DDM) in a multiple-input multiple-output (MIMO) radar network, the method comprising the following steps: - generating and transmitting a coded signal by means of a Doppler multiplexing method (DDM) to at least a first radar (Rad1) and a second radar (Rad2) in a MIMO radar network, wherein each radar (Rad1, Rad2) comprises at least one transmitter and a receiver, and wherein the coded signal has a corresponding DDM code (DC1, DC2) which unambiguously identifies each radar (Rad1, Rad2); - receiving and preprocessing the reflected signals for digital processing and conversion into a range-Doppler velocity-matrix; - performing a threshold detection for each range-Doppler matrix to determine the cells with potential signals of the corresponding radar (Rad1, Rad2), wherein the threshold detection performs a binary transformation on these signals; - convolving a corresponding latent binary signal with each of the explicit DDM codes and assigning the corresponding signal explicitly to one of the radars (Rad1, Rad2), - determining in the bistatic path (BP) a clearly assigned signal, whose signal parameters overlap for each radar, and in the monostatic path (MP1, MP2) a clearly assigned signal for each radar (Rad1, Rad2), wherein the signal of the corresponding monostatic path (MP1, MP2) differs from the signal of the bistatic path (BP); - determining a signal which is clearly assigned both in the bistatic path (BP) and in each monostatic path (MP1, MP2), wherein the signal of the monostatic path (MP1, MP2) is determined starting from the corresponding signal of the bistatic path (BP), and - MIMO angle estimation using the complete aperture of a virtual MIMO array formed by signals that are clearly assigned in all paths.

2. The method according to claim 1, further comprising the steps of: - determining the signals that have been assigned only in the bistatic path (BP); - performing MIMO angle estimation using the bistatic aperture of the first and second radars (Rad1, Rad2), and / or additionally comprising the following steps: - determining signals which have been assigned only in one of the monostatic paths (MP1, MP2); - using at least one of the monostatic aperture of the first radar (Rad1) and the monostatic aperture of the second radar (Rad2) for MIMO angle estimation.

3. The method according to any one of the preceding claims, wherein: Each radar comprises a plurality of transmitters (Tx), and the corresponding DDM code comprises an explicit phase shift of the signal of each transmitter (Tx) of the corresponding radar (Rad1, Rad2).

4. The method according to any one of the preceding claims, wherein: Using a point-symmetrical search, starting from the associated signal of the bistatic path (BP), the signal which is clearly assigned both in the bistatic path (BP) and in each monostatic path (MP1, MP2) is determined.

5. The method according to claim 4, wherein: The search range of the point-symmetrical search is respectively expanded by one cell for each dimension from the zero cell up to the maximum possible deviation.

6. The method according to any one of the preceding claims, wherein: The method is carried out with the aid of a linear frequency modulated chirp signal or an exponential frequency modulated chirp signal or a stepwise frequency modulated chirp signal as the sequential transmission signal.

7. The method according to any one of the preceding claims, wherein: The method is implemented according to a time multiplexing mode (TDM), and the corresponding DDM codes (DC1, DC2) of the first radar (Rad1) and the second radar (Rad2) are exchanged within a first time window and a second time window (Z1, Z2), and the method is implemented for the first time window (Z1) and the second time window (Z2), respectively; In this case, the results are compared with one another and only identical signals are taken into account for each time window ( Z1 , Z2 ).

8. The method according to claim 7, wherein: The results of the first time window and the second time window (Z1, Z2) are compared with each other before the step of performing the MIMO angle estimation, and the MIMO angle estimation is performed only for the considered signal.

9. The method according to any one of the preceding claims 1 to 6, wherein: The radar network includes more than two radars, and wherein each of the DDM codes unambiguously identifies the corresponding radar.

10. The method according to any one of claims 7 or 8 in combination with claim 9, wherein: The DDM codes of the individual radars are exchanged in pairs for each of the two time windows ( Z1 , Z2 ).

11. A radar network having at least a first and a second radar, in which the method according to one of claims 1 to 10 is implemented.

Images