Method for generating simulated radar return signals and radar target simulator

Abstract

The invention relates to a method for generating a simulated radar echo signal and a radar target simulator. The method according to the invention has the following method steps: transmitting a radar signal by a radar sensor onto a radar target simulator; receiving the radar signal in the radar target simulator; filtering the radar signal in the radar target simulator by means of a low-pass filter having a known filter curve; calculating a corrected frequency spectrum by means of a correction of a previously determined frequency spectrum by means of the filter curve; calculating a power of the radar signal corresponding to the corrected frequency spectrum; calculating a scaled radar signal from the filtered radar signal; calculating a radar echo signal as a reflection of the scaled radar signal; and transmitting the radar echo signal by a transmitting antenna of the radar target simulator onto the radar sensor to be tested. This makes it possible for the method to be applied as error-free as possible even in the case of a radar target simulator having a small bandwidth.

Description

TECHNICAL FIELD

[0001] The invention relates to a method for generating a simulated radar echo signal, comprising the following method steps: transmitting a radar signal by a radar sensor to be tested onto a receiving antenna of a radar target simulator; receiving the radar signal in the radar target simulator; filtering the radar signal in the radar target simulator by means of a low-pass filter; and transmitting the simulated radar echo signal by a transmitting antenna of the radar target simulator onto the radar sensor to be tested. Furthermore, the invention relates to a radar target simulator for generating a simulated radar echo signal. BACKGROUND

[0002] For the exact detection of driving situations, automated vehicles require radar sensors with high resolution, which provide a broad field of view in addition to a detailed image and also height, distance and speed information. For testing these sensors, radar target simulators are applied, i.e. test benches, which detect the radar signals of a vehicle radar system installed as a specimen, calculate the radar echo of the signals on the basis of a real-time model and generate a delayed response signal corresponding to the calculated echo and emit it onto the specimen. The detection of physical targets is tested or rather preformed by the specimen in this way. The radar signals from the specimen pass through the target simulator on measurement through a low-pass filter, which serves as an anti-aliasing filter. Such a signal filter reduces the bandwidth of the measurement, since such a signal filter only lets the signal pass unattenuated over a limited frequency spectrum. In the transition region at the border of the adjacent filter spectrum, the low-pass filter attenuates the signal strength.

[0003] However, such low-pass filters are required to avoid or reduce the aliasing effect, which occurs in signal processing when a signal is digitized: the original signal is sampled at regular time intervals here and in the subsequent reproduction is established again by means of an analog low-pass filter. In order that the original signal can be correctly established again, it must be sampled according to the Nyquist sampling theorem at a sampling rate which is greater than twice the highest frequency occurring in the signal. If the sampling theorem is violated due to too low a sampling rate, the original high-frequency components above half the sampling rate (Nyquist frequency) are interpreted as lower frequencies, since undersampling occurs for this. This undesirable phenomenon is called aliasing. In digital signal processing, so-called pre-filtering is therefore used to avoid the aliasing effect. An analog low-pass filter is applied to the signal here before digitization. The frequencies of the signal above the Nyquist frequency are thus attenuated by this. Such electronic filters are preferably as steep-edged as possible, which can be achieved by expensive filters of higher order. Even so, the components of the signal below the Nyquist frequency are attenuated and the components above the Nyquist frequency are not completely eliminated. The exact choice of the cut-off frequency is therefore a compromise in practice between eliminating the aliasing effect and obtaining a useful signal. However, it is not possible to obtain a useful signal completely in the case of the application of such filters.

[0004] If the radar sensor is now to be verified by means of a target simulator, the target simulator must detect the entire bandwidth of the signal in order to calculate the radar reflection area (or radar scattering cross-section, ) correctly in the radar. In order to avoid errors in connection with this, radar sensors have hitherto only been tested with target simulators which can provide the required bandwidth. If the radar sensor emits a signal with a bandwidth of 2 GHz, for example, approximately half the power - depending on the signal form and the filter curve of the target simulator - is lost in the case of the application of a target simulator with a bandwidth of 1 GHz, so that the reflection area of the simulated target from the radar's point of view is approximately 3 dB smaller than originally set in the target simulator.

[0005] A higher bandwidth is a frequently occurring, but to some extent difficult to realize, desire. This often requires more expensive components, in particular expensive processors, as long as they can be used in total for the required bandwidth determined by the sampling rate. Wideband components in the high-frequency field are difficult to construct, so that they are difficult to use and very expensive. If the processing takes place digitally, the processor is in most cases decisive for the sampling rate and thus finally for the required bandwidth. SUMMARY

[0006] On the basis thereof, the object of the present application is to provide a method for generating simulated radar echo signals, which can be applied as error-free as possible even in the case of radar target simulators having a small bandwidth.

[0007] The object is solved by the content of the independent claims. Preferred refinements emerge from the dependent claims.

[0008] According to the application, this is provided by a method for generating simulated radar echo signals, which has the following method steps:

[0009] a radar signal having a known bandwidth is transmitted by the radar sensor to be tested onto a receiving antenna of a radar target simulator;

[0010] the radar signal is received in the radar target simulator having the receiving antenna;

[0011] the radar signal is filtered in the radar target simulator by means of a low-pass filter having a known filter curve;

[0012] the frequency spectrum of the filtered radar signal is determined over the entire bandwidth given by the filter curve of the low-pass filter;

[0013] a corrected frequency spectrum is calculated by correction of the previously determined frequency spectrum by means of the filter curve, or in other words a corrected frequency spectrum is calculated which is obtained by correction of the previously determined frequency spectrum by means of the filter curve, so that as corrected frequency spectrum a frequency spectrum is obtained which corresponds to the frequency spectrum of the transmitted unfiltered radar signal;

[0014] the power of the radar signal corresponding to the corrected frequency spectrum is calculated;

[0015] a scaled radar signal is calculated from the filtered radar signal, wherein the power of the scaled radar signal is equal to the power of the radar signal corresponding to the corrected frequency spectrum, in such a way that the amplitudes of the filtered radar signal are subjected to a scalar multiplication which is identical for all frequencies over the entire frequency spectrum of the filtered radar signal;

[0016] the radar echo signal is calculated as a reflection of the scaled radar signal; and

[0017] the radar echo signal is transmitted by a transmitting antenna of the radar target simulator onto the radar sensor to be tested.

[0018] The term "radar sensor" denotes a unit for transmitting and receiving radar signals (with an antenna and other means required for this), as is customary in sensors for providing driver assistance systems in the case of motor vehicles. The filter curve of a filter is to be understood as the attenuation coefficient diagram for signals passing through the filter with respect to the frequency of the signals, i.e. in particular the amplitude response or the intensity response (squared amplitude response) of the signals. In general, the filter curve comprises a passband, which is characterized in that frequency portions of the signals lying in the passband pass through the filter unattenuated, and at least one transition band adjoining the passband, which is characterized in that frequency portions of the signals lying in the transition band pass through the filter in attenuated form. The bandwidth provided by the filter curve of a low-pass filter is to be understood as the totality of frequencies covered by the passband and the transition band of the low-pass filter, i.e. all frequencies for which the filter curve of the low-pass filter gives an attenuation coefficient greater than zero.

[0019] The following step is carried out by taking into account the filter-based attenuation known on the basis of the filter curve: the corrected frequency spectrum is calculated by means of a correction of the previously determined frequency spectrum by means of the filter curve, so that a frequency spectrum is obtained as the corrected frequency spectrum which corresponds to the frequency spectrum of the transmitted unfiltered radar signal: since it is known on the basis of the filter curve how much attenuation there is on the basis of the filter for each frequency, it is possible in this way to infer from the attenuated amplitude of the transmitted radar signal its unattenuated amplitude.

[0020] By means of the application it is possible to manipulate the signal in such a way that the application of a radar target simulator with a smaller bandwidth than the radar sensor under test also contributes to the correct determination of the radar reflection surface in the radar sensor. By means of the application, a conventional radar target simulator, which is correspondingly improved in terms of the method, can also test radar sensors with a higher bandwidth in terms of the correct determination of the radar reflection surface. Since the filter curve of the low-pass filter of the radar target simulator is known, it is possible to determine on the basis of the known bandwidth of the radar sensor how large the lost power portion is. By means of this information it is possible to emit more power in terms of the radar target simulator in order to compensate for the lost power portion. As a result, the determination of the reflection surface or scattering cross section in the radar sensor provides the desired result, even if the signal is limited in terms of bandwidth. The advantage of the method according to the application is therefore that a change in hardware in the radar target simulator is not necessary. In particular, the radar target simulator is operable without a particularly high-performance processor, which provides a sufficiently high sampling rate for processing very wideband signals in real time. The application thus enables a quick improvement in terms of bandwidth and a large cost saving.

[0021] In principle, the transmitting antenna can be identical to the receiving antenna of the radar target simulator. Preferably, however, it is provided that the transmitting antenna and the receiving antenna of the radar target simulator are constituted as separate, mutually separated devices. It is preferably additionally provided here that the transmitting antenna is moved relative to the radar sensor to be tested. This enables the simulation of radar signals reflected by virtual objects which are not only located directly in front of the radar sensor to be tested, but also on the side thereof.

[0022] Furthermore, it is preferably provided in accordance with the application that the low-pass filter is an anti-aliasing filter having a cut-off frequency which does not enable sampling of the transmitted radar signal over its entire bandwidth. This means that a conventional hardware is applied.

[0023] It is furthermore provided in accordance with the application that a radar target simulator for generating a simulated radar echo signal has:

[0024] a receiving antenna for receiving a radar signal in the radar target simulator;

[0025] a low-pass filter having a known filter curve for filtering the radar signal;

[0026] a computing unit, which is set up for determining the frequency spectrum of the filtered radar signal over the entire bandwidth given by the filter curve of the low-pass filter, calculating a corrected frequency spectrum by correction of the previously determined frequency spectrum by means of the filter curve, such that as the corrected frequency spectrum a frequency spectrum is obtained which corresponds to the frequency spectrum of the transmitted unfiltered radar signal, calculating the power of the radar signal corresponding to the corrected frequency spectrum, calculating a scaled radar signal from the filtered radar signal, wherein the power of the scaled radar signal is equal to the power of the radar signal corresponding to the corrected frequency spectrum in such a way that the amplitudes of the filtered radar signal are subjected to a scalar multiplication which is identical for all frequencies over the entire frequency spectrum of the filtered radar signal, calculating the simulated radar echo signal as a reflection of the scaled radar signal, and

[0027] a transmitting antenna for transmitting the simulated radar echo signal.

[0028] The preferred refinements of the radar target simulator result analogously to the above-described preferred refinements of the method in accordance with the application. BRIEF DESCRIPTION OF DRAWINGS

[0029] The application is explained in further detail below on the basis of preferred embodiments with reference to the drawings. The drawings show:

[0030] Figure 1 a radar target simulator in accordance with one embodiment of the application is shown schematically together with a radar sensor to be tested; and

[0031] Figure 2 The method according to an embodiment of the application is schematically illustrated. DETAILED DESCRIPTION

[0032] Figure 1 A radar target simulator 1 for generating simulated radar echo signals is schematically illustrated, comprising a receiving antenna 2 for receiving a radar signal in the radar target simulator 1, a low-pass filter 3, a computing unit 4 and a transmitting antenna 5. The radar sensor 6 to be tested is arranged with respect to the radar target simulator 1 such that the radar signal emitted by the radar sensor 6 is receivable by the receiving antenna 2 of the radar target simulator 1. As indicated by the two arrows, the radar sensor 6 to be tested is arranged in a stationary manner with respect to the radar target simulator 1. Figure 1 It can be seen that the transmitting antenna 5 and the receiving antenna 2 of the radar target simulator 1 are configured as separate, mutually separated devices. The receiving antenna 2 of the radar target simulator 1 is stationary, while the transmitting antenna 5, as indicated by the two arrows, is movable, i.e. can be moved in a circular manner approximately 180° around the radar sensor 6 to be tested. This enables the simulation of radar signals reflected by virtual objects not only exactly in front of the radar sensor 6 to be tested, but also laterally thereto.

[0033] The low-pass filter 3 is an anti-aliasing filter having a known filter curve, but said anti-aliasing filter has a cut-off frequency which does not enable sampling of the radar signal transmitted by the radar sensor 6 to be tested over its entire bandwidth. This would in principle lead to the generation of false simulated radar echo signals in the conventional operation of the aforementioned radar target simulator 1 for the reasons already mentioned above. Therefore, according to an embodiment of the application is a method for generating simulated radar echo signals, said method having the following method steps and being illustrated schematically in Figure 2

[0034] In a first step S1, a radar signal having a known bandwidth is transmitted by the radar sensor 6 to be tested onto the receiving antenna 2 of the radar target simulator 1. At the radar target simulator, the radar signal is received in a step S2 with the receiving antenna 2. In a subsequent step S3, the radar signal is filtered in the radar target simulator 1 by means of the low-pass filter 3 at the radar target simulator in a known manner in order to avoid or reduce aliasing effects. Up to this point, the method described here basically corresponds to the conventional method for operating the radar target simulator.

[0035] ​In step S4, the spectrum of the filtered radar signal over the entire bandwidth of the low-pass filter 3 given by the filter curve is now determined. In step S5, a corrected spectrum is then calculated by a correction of the previously determined spectrum by means of the filter curve, such that as corrected spectrum a spectrum is obtained which corresponds to the spectrum of the transmitted unfiltered radar signal. In step S6, the power of the radar signal corresponding to the corrected spectrum is then calculated. In step S7, a scaled radar signal is then calculated from the filtered radar signal, wherein the power of the scaled radar signal is equal to the power of the radar signal corresponding to the corrected spectrum, by subjecting the amplitudes of the filtered radar signal over its entire spectrum to a scalar multiplication which is identical for all frequencies. Finally, in step S8, a radar echo signal is calculated on the basis of the scaled radar signal and transmitted from the transmitting antenna 5 of the radar target simulator 1 onto the radar sensor 6 to be tested. The radar echo signal is calculated as a reflection of the scaled radar signal. The emitted radar echo is thus an artificial imitation of a radar echo of the scaled radar signal, which is not received by the receiving antenna 2.

[0036] By taking into account the filter attenuation known on the basis of the filter curve, the following step is carried out, namely the calculation of a corrected spectrum by a correction of the previously determined spectrum by means of the filter curve, such that the spectrum obtained corresponds to the spectrum of the transmitted unfiltered radar signal: since it is known on the basis of the filter curve how much the attenuation is based on the filter for each frequency, the unattenuated amplitude of each frequency portion of the transmitted radar signal can be inferred from its attenuated amplitude in this way. In this way, the signal can be manipulated such that, as described above, the application of a radar target simulator with a smaller bandwidth than the radar sensor also contributes to the correct determination of the radar reflection surface in the radar. Thereby, with the radar target simulator 1 which is otherwise conventional, which is improved by the aforementioned method, radar sensors with a higher bandwidth can also be tested in terms of the determination of the radar reflection surface. The only prerequisite for this is that the filter curve of the target simulator 1 is known and that the transition region of the filter covers a sufficiently large spectrum. Thus, on the basis of the known bandwidth of the radar sensor, it can be determined how large the lost power portion is, so that more power can be emitted by the target simulator 1 by means of the transmitting antenna 5 in order to compensate for the lost power portion. Overall, the reflection surface or the backscatter cross section which is indirectly determined in this way in the radar sensor contributes to the desired (correct) result, even if the signal in the radar target simulator 1 is limited in terms of bandwidth.

[0037] List of reference signs

[0038] 1 radar target simulator

[0039] 2 receive antenna

[0040] 3 low-pass filter

[0041] 4 computing unit

[0042] 5 transmit antenna

[0043] 6 radar sensor to be tested

[0044] S1 transmitting a radar signal

[0045] S2 receiving a radar signal

[0046] S3 filtering the radar signal

[0047] S4 determining a frequency spectrum of the filtered radar signal

[0048] S5 computing a corrected frequency spectrum

[0049] S6 computing a power of the radar signal corresponding to the corrected frequency spectrum

[0050] S7 computing a scaled radar signal

[0051] S8 transmitting the scaled radar signal as an analog radar echo signal

Claims

1. A method for generating a simulated radar echo signal, the method comprising the following method steps: - transmitting a radar signal having a known bandwidth by a radar sensor (6) to be tested onto a receiving antenna (2) of a radar target simulator (1); - receiving the radar signal in the radar target simulator having the receiving antenna (2); - filtering the radar signal in the radar target simulator (1) by means of a low-pass filter (3) having a known filter curve; - determining a spectrum of the filtered radar signal over the entire bandwidth given by the filter curve of the low-pass filter (3); - calculating a corrected spectrum by a correction of the previously determined spectrum by means of the filter curve, so that a spectrum is obtained as the corrected spectrum which corresponds to the spectrum of the transmitted unfiltered radar signal; - calculating a power of the radar signal corresponding to the corrected spectrum; - calculating a scaled radar signal from the filtered radar signal, wherein the power of the scaled radar signal is equal to the power of the radar signal corresponding to the corrected spectrum in such a way that the amplitudes of the filtered radar signal are subjected to a scalar multiplication which is identical for all frequencies over the entire spectrum of the filtered radar signal; - calculating a radar echo signal as a reflection of the scaled radar signal; and - transmitting the radar echo signal by a transmitting antenna (5) of the radar target simulator (1) onto the radar sensor (6) to be tested. The receiving antenna (2) and the transmitting antenna (5) of the radar target simulator (1) are constituted as separate, mutually separated devices. The transmitting antenna (5) is moved relative to the radar sensor to be tested. The low-pass filter (3) is an anti-aliasing filter having a cut-off frequency which does not enable a sampling of the transmitted radar signal over its entire bandwidth.

5. A radar target simulator (1) for generating a simulated radar echo signal, the radar target simulator having: - a receiving antenna (2) for receiving a radar signal in the radar target simulator (1); - a low-pass filter (3) having a known filter curve for filtering the radar signal; - a calculating unit (4) which is set up for: - determining a spectrum of the filtered radar signal over the entire bandwidth given by the filter curve of the low-pass filter (3); - calculating a corrected spectrum by a correction of the previously determined spectrum by means of the filter curve, so that a spectrum is obtained as the corrected spectrum which corresponds to the spectrum of the transmitted unfiltered radar signal; - calculating a power of the radar signal corresponding to the corrected spectrum; - calculating a scaled radar signal from the filtered radar signal, wherein the power of the scaled radar signal is equal to the power of the radar signal corresponding to the corrected spectrum in such a way that the amplitudes of the filtered radar signal are subjected to a scalar multiplication which is identical for all frequencies over the entire spectrum of the filtered radar signal; and - calculating a simulated radar echo signal as a reflection of the scaled radar signal; and - a transmitting antenna (5) for transmitting the simulated radar echo signal. The transmitting antenna (5) and the receiving antenna (2) of the radar target simulator (1) are constituted as separate, mutually separated devices. The transmitting antenna (5) is movable. The transmitting antenna (5) is movable. ​ ​ 2. The method of claim 1, wherein, ​ 3. The method of claim 2, wherein, ​ 4. The method according to one of the preceding claims, wherein ​ ​ ​ ​ ​ ​ ​ ​ - calculating a scaled radar signal from the filtered radar signal, wherein, ​ ​ ​ 6. The radar target simulator of claim 5, wherein, ​ 7. The radar target simulator of claim 6, wherein, ​ 8. The radar target simulator according to one of claims 5 to 7, wherein, The low-pass filter (3) is an anti-aliasing filter with a cut-off frequency that does not enable sampling of the transmitted radar signal over its entire bandwidth.

Images