Method and device for detecting interference in a radar signal

Abstract

The invention relates to a method for detecting interference in a radar signal. One step comprises transmitting a frequency ramp. Another step comprises receiving the radar signal in the time domain. An IQ mixer is applied to the received radar signal and separates the radar signal into a wanted signal and a non-wanted signal part. Another step comprises identifying one or more interference regions in the non-wanted signal part in the time domain and / or the frequency domain. An interference region is identified if the absolute value of the amplitude of the non-wanted signal part or the change of the amplitude of the non-wanted signal part exceeds a predefined threshold. The invention also relates to a device for detecting interference in a radar signal.

Description

Technical Field

[0001] The present invention relates to a method for detecting interference in radar signals and a device for detecting interference in radar signals. Background Technology

[0002] To assist vehicle drivers and to achieve autonomous driving, radar sensors are increasingly being relied upon. This typically involves continuous wave radar systems operating in the 76 GHz band. Particularly popular are FMCW ("Frequency Modulated Continuous Wave") modulation and its further developments and variations. Here, the frequency of the transmitted radar waves is periodically modulated. The transition from the lowest to the highest frequency is called a frequency ramp, signal sequence, or short burst or chirp. Besides frequency offset, i.e., the difference between the lowest and highest frequencies of the corresponding FMCW ramp, a radar system is also characterized by the frequency ramp, which is limited by the receiver's bandwidth.

[0003] The more vehicles on the road equipped with radar sensors, the greater the risk of unwanted, interfering cross-influences or interference effects. In this type of interference, commonly referred to as interference, the slope of the native FMCW ramp is generally different from the slope of the interfering FMCW ramp. Therefore, an interfering signal is generated in the baseband, with its frequency continuously increasing from the negative highest baseband frequency to the positive highest baseband frequency, or vice versa. The corresponding time period is defined by the difference in the slopes of the native and interfering ramps, as well as by the receiver's sampling rate and receiving bandwidth.

[0004] To reduce interference effects, a radar sensor with a controllable on or off switch is known from document DE 10 2014 112 806 A1, which attenuates or interrupts the transmission of signals to the radar sensor's transmitting antenna.

[0005] German patent document DE 10 2018 200 753 A1 discloses a method for correcting interfered radar signals. First, the dominant peak is determined in the spectrum of the received radar signal. Then, an auxiliary signal is obtained by removing a portion of the dominant peak from the radar signal, and the interference region of the radar signal is identified using the auxiliary signal. Subsequently, the corrected radar signal is generated by interpolating the radar signal within the identified interference region using the determined dominant peak. Summary of the Invention

[0006] This invention provides a method for determining the interference region without the need for auxiliary signals.

[0007] This invention provides a method for detecting interference in radar signals according to the invention, and an apparatus for detecting interference in radar signals according to the invention. Preferred configurations of the invention are the subject matter described in the accompanying drawings and embodiments.

[0008] The method for detecting interference in radar signals according to the invention includes the step of transmitting a frequency ramp. Here, in particular, a linear FMCW ramp (see above) is transmitted, wherein, for example, a narrow-band signal is modulated onto the carrier frequency. For this purpose, an FMCW radar device, including a local oscillator and an antenna, can be used, for example.

[0009] In another step, the radar signal in the time domain is received. The received radar signal may, for example, include portions of the emitted frequency ramp reflected by multiple targets. Furthermore, the received radar signal may have interference from other transmitters.

[0010] According to the present invention, an IQ mixer is applied to a received radar signal to separate the radar signal into a useful signal and a useless signal portion. The basic operating principle of an IQ mixer is well known from communication technology. In this case, it is used in particular to distinguish between “positive” and “negative” frequencies (i.e., whether the received frequency is greater than or less than the current frequency of the local oscillator when downmixed to baseband). Therefore, EP 3 740 777 B1 also implicitly describes a method without an IQ mixer, in which the negative portion of the spectrum is the complex conjugate of the positive portion (i.e., mirrored at 0 Hz, having the same magnitude but with reversed phase) – hence the figures there are limited to the positive portion of the spectrum.

[0011] For a rising FMCW ramp, the runtime of the reflected signal results in the local oscillator's current transmitted signal already having a higher frequency than any useful signal component of the signal mixture in the received signal. For a falling FMCW ramp, the opposite is true; all useful signal components have higher frequencies. However, system noise can generally be assumed to be "white noise," meaning it has the same power density across all frequencies. Therefore, ideally, half the power of the system noise falls into the sidebands with no useful power.

[0012] In the prior art, sidebands with no useful power are simply ignored. According to the present invention, it is examined whether sidebands with no useful power contain more power than expected, compared to the power expected based on system noise.

[0013] This method enables the identification of one or more interfering regions within a segment devoid of useful signal in the time and / or frequency domains (ideally containing only unavoidable system noise, and thus, in particular, thermal noise, without such interference). Interfering time samples (Zeitsamples) are identified if the absolute value of the amplitude of the useless signal segment of the interfering time sample, or the change in amplitude (first derivative) of the useless signal segment, exceeds a pre-defined threshold.

[0014] When determining the threshold, it can be assumed that the radar sensor is undisturbed for most of the time. However, for a particular radar sensor, the system noise varies primarily due to changes in its operating temperature, and therefore changes only relatively slowly compared to the typically short measurement period of 50 ms. Accordingly, this comparison with the currently expected power based on the system noise can be performed using very simple methods, such as a time average plus a 3 sigma limit, but it can also be done using robust statistical data such as the median or quantiles.

[0015] Time samples that are detected as being interfered with and are close to each other, with a small number (e.g., 5) of sampling intervals, are grouped into a single interference region belonging to the same whole. Figure 2 For time samples that are far apart from each other and are detected as being interfered with, additional interference areas are generated.

[0016] Preferably, the step of identifying one or more interference regions in the portion without a useful signal is performed only if the total power obtained in the portion without a useful signal is equal to or greater than a limit value for the total power. For example, if the received signal contains interference from other transmitters, the total power contained therein may be significantly greater than the expected total power that would ideally occur due to unavoidable system noise (i.e., the expected total power in the portion without a useful signal is very small). Therefore, by obtaining the total power in the portion without a useful signal, it is possible to determine whether an interfering transmitter is present.

[0017] The determination of the limiting values ​​used for comparison can again be done using simple methods, such as the time average plus the 3 sigma limit—but it can also be done using robust statistical data, such as the median or quantiles.

[0018] Preferably, the interference region in the useful signal can be determined based on symmetry, according to the interference region in the part without useful signal, and then the corrected radar signal can be generated according to EP 3 740 777 B1.

[0019] However, the total power in the portion without a useful signal can also increase due to reasons other than interference. Another reason could be overload on the receiving side, such as when a highly reflective object is located at a very small distance. As a result, the received signal sets at more power than the radar sensor's limited operating range. This leads to nonlinearity in the power amplifier and clipping in the analog-to-digital converter. Accordingly, the total power in the portion without a useful signal increases significantly over the entire duration (and all chirps).

[0020] An apparatus for detecting interference in radar signals includes: a transmitting device for transmitting a frequency ramp; a receiving device for receiving radar signals in the time domain; an IQ mixer configured to separate the received radar signals into useful signal and useless signal portions; and a computing device for identifying one or more interference regions within the useless signal portions in the time and / or frequency domains. The computing device is configured to identify an interference region if the absolute value of the amplitude of the useless signal portion or the variation in the amplitude of the useless signal portion exceeds a predetermined threshold. Attached Figure Description

[0021] The invention will now be explained in more detail with reference to the embodiments shown in the schematic diagrams. Herein lies: Figure 1 A block diagram illustrating an embodiment of the method according to the present invention is shown; Figure 2 The diagram illustrates the separation of received radar signals into useful and useless components; and Figure 3 The illustration shows an example of the FMCW ramp and the time curves of reflected and interfering signals in the received signal when using an IQ mixer.

[0022] The accompanying drawings are intended to provide a further understanding of embodiments of the invention. They illustrate the embodiments and, in conjunction with the description, serve to explain the principles and concepts of the invention. Many advantages, among other embodiments and mentioned benefits, are derived from the accompanying drawings. Elements in the drawings are not necessarily shown to scale correctly.

[0023] In the accompanying drawings, identical, functional, and operational elements, features, and components are given the same reference numerals, unless otherwise specified. Detailed Implementation

[0024] Figure 1A block diagram illustrating an embodiment of the method according to the invention is shown. After a transmit frequency ramp, a radar signal (received signal RF) in the time domain is received by receiving antenna 1 and input together with a reference signal SK from local oscillator 3 into IQ mixer 2. IQ mixer 2 generates a complex baseband signal. Subsequent IQ-immbalance correction can be applied to compensate for mixer errors and thus improve the purity of the portion without a useful signal.

[0025] Next, the complex baseband signal is separated into a useful signal SN and a portion SS without a useful signal in function block 4. This can be achieved, for example, directly after the FFT or according to filtering. The useful signal SN is input to function block 5, and the portion SS without a useful signal is input to function block 6.

[0026] In function block 6, interference or saturation detection is performed on the portion of SS without a useful signal, and its location is output to function block 5. In function block 5, interference suppression can be performed on the useful signal SN. Furthermore, in function block 6, the total power P of the portion without a useful signal is calculated, and it is passed to function block 8 for further evaluation.

[0027] In functional block 7, actual radar signal processing is performed to determine radar reflection 9 and its characteristics. In functional block 8, system reactions can be performed based on the calculated power P, such as reduced gain in the receiver path due to saturation, loss of reporting sensitivity, and reduced line-of-sight.

[0028] Figure 2 The diagram illustrates the separation of the received radar signal into the useful signal (SN) and the non-useful signal portion (SS). For simplicity, only the time curve of the real part of the signal is shown. This is in Figure 1 It occurs in function block 4.

[0029] Actual interference detection in Figure 1 This is performed in function block 6. For this purpose, the magnitude of the complex portion of SS without a useful signal is formed and compared with the threshold TSchwelle, resulting in the signal SS_POS. Figure 2 (Example with an interference area). The signal SS_POS is then transmitted to the location (SN_POS) in the useful signal, which must be considered separately for each individual interference area: Due to the symmetrical receiving bandwidth, the location of interference in the useful signal is approximately a mirror image of its location in the signal without a useful signal. The location of the interference in the useful signal is the detected interference located in the signal without a useful signal. Figure 2 Whether it is on the left or right depends on its own emission chirp (CTX) and interference chirp (CI) in the time-frequency plane. Figure 3 The sign of the angle in the curve can be determined, for example, based on the shape of the interference curve. This may be different for each individual interference region. The axis of symmetry is the side with the low-frequency component. For simplicity, the following hypothesis can also be made by determining the power of the useful signal SN on the left and right sides of the interference region: the region containing more power is the interference region being sought.

[0030] Figure 3 The illustration shows an example of the FMCW ramp and the time curves of reflected and interfering signals in the received signal when using an IQ mixer.

[0031] The FMCW radar transmits a frequency ramp (chirp) CTX, which may be reflected by targets in the environment. The frequency ramp CTX... Figure 3 The reflected slope CR (received chirp) is represented by a thick solid line. Figure 3 The solid line appears below the frequency ramp CTX emitted from the signal.

[0032] The brackets to the left of the frequency axis indicate the sensitivity range RXR of the radar receiver. The receiver is constructed such that the current transmit frequency of local oscillator 3 is always subtracted from the received signal CR. This process is called downmixing to baseband. Since the current transmit frequency of oscillator 3 has become larger in our example, a larger frequency is subtracted from the received chirp CR, resulting in a negative frequency.

[0033] In the time domain, this is a sinusoidal signal. The distance between the radar sensor and the target can be calculated from the constant frequency difference.

[0034] exist Figure 2 The left side of the middle section shows the real part of an exemplary received signal in the time domain.

[0035] If an interference source (e.g., from another FMCW radar) crosses the radar sensor's receive bandwidth RXR and transmits an interfering chirp CI (see... Figure 3 If the intersecting straight lines (from the top left to the bottom right) are mixed, their power will also be down-mixed. During the period when the interference chirp CI is within the receiving bandwidth RXR, this manifests as interference in the time-domain signal; see [reference needed]. Figure 2 This interference is also known as interference. Since the interfering chirp CI passes through all frequencies of the baseband, it causes an increase in noise at those frequencies in the frequency domain—and thus a decrease in sensitivity. This results in a separate interference region in the received signal. If multiple interfering chirps occur during the duration of the transmitted chirp (e.g., due to other radars in the environment), this will result in multiple interference regions in the received signal.

[0036] The receive bandwidth RXR is divided into negative frequency range and positive frequency range. Figure 3 In the image, this area is outlined by two dashed lines.

[0037] According to the present invention, a complex-valued mixer (IQ mixer) is used. Therefore, an "image rejection filter" (see [reference]) can be used. Figure 1 Box 4) Removes unwanted sidebands and their power. The suppressed region R is thus located in... Figure 3 It is presented in the form of shadows.

[0038] According to the present invention, this removed sideband is used to determine the presence of interference. The useful signal (including the presence of interference or saturation and the target) is forwarded to further radar signal processing as in conventional schemes (see...). Figure 1 (Box 7 in the text). The signal SS, which contains no useful signal, only contains interference or saturation, therefore (in...) Figure 1 Box 6) is used to detect interference or saturation. Here, it can be determined by... Figure 1 Box 6 in the middle outputs the location X of the disturbance or saturation. Figure 1 Box 5. The determination of position X is also shown in, for example, in... Figure 2 middle.

[0039] The region DS on the time axis gives the time range when the interference chirp CI crosses the receive bandwidth RXR. Due to its symmetry, the interference chirp (CI) lies within the receive bandwidth for the same duration on both sides.

[0040] In this invention, multiple features are designated with the names "first" and "second". These names are used only to clearly distinguish the various features. In particular, no spatial or functional arrangement or priority should be derived from them.

[0041] If the list of alternatives in this application is named "or", it should be understood not only that each of the listed alternatives is used individually, but also, where reasonable, that a combination of multiple or all of the listed alternatives.

Claims

1. A method for detecting interference in radar signals, the method comprising the following steps: Transmission frequency ramp; Receive radar signals in the time domain; An IQ mixer is applied to the received radar signal and the radar signal is separated into a useful signal and a useless signal portion; Identify one or more interference regions in the portion of the signal-free area in the time and / or frequency domains, wherein... An interference region is identified if the absolute value of the amplitude of the portion without useful signal or the change in the amplitude of the portion without useful signal exceeds a pre-defined threshold.

2. The method according to claim 1, wherein, The interference region is defined as the time domain of points within it that are close to each other, at which point the amplitude of the portion without useful signal or the change in the amplitude of the portion without useful signal exceeds a predetermined threshold.

3. The method according to claim 2, wherein, For time points that are far apart from each other and are detected as being interfered with, find out the other interference areas.

4. The method according to any one of the preceding claims, wherein: The step of identifying one or more interference regions in the portion of the region without useful signals is performed only if the total power obtained in the portion without useful signals is equal to or greater than the limit value of the total power.

5. The method according to claim 4, further comprising: A warning message is output when the total power is equal to or greater than the limit value of the total power.

6. The method according to any one of the preceding claims, further comprising: Based on symmetry, the interference region in the useful signal is determined according to the relevant interference region in the part without useful signal.

7. The method according to any one of the preceding claims, further comprising: The corrected radar signal is generated by interpolating the useful signal in the identified interference area.

8. An apparatus for detecting interference in radar signals, the apparatus comprising: The transmitting device is used to transmit the frequency ramp; A receiving device for receiving radar signals in the time domain; An IQ mixer, configured to separate a received radar signal into a useful signal and a useless signal portion; as well as A computing device for identifying one or more interference regions in the portion of the absence of useful signal in the time domain and / or frequency domain, wherein the computing device is configured to: identify an interference region if the absolute value of the amplitude of the portion of the absence of useful signal or the change in the amplitude of the portion of the absence of useful signal exceeds a predetermined threshold.

Images