FMCW radar with frequency-hopping technique

Abstract

A method comprising: generating a frequency-modulated local oscillator signal (sLO(t)) with a frame partitioned into N chirp positions, where N is a predetermined integer; wherein the frame contains a scan chirp at one or more chirp positions and regular chirps at the remaining chirp positions, the regular chirps having a chirp bandwidth and a start frequency (fSTART), and the at least one scan chirp having a scan bandwidth greater than the chirp bandwidth; transmitting an RF signal representing the frame, wherein the transmit power is zero or reduced to less than the power of a regular chirp during the transmission of the at least one scan chirp;Receiving an RF radar signal (yHF(t)) corresponding to the frame, down-converting the received RF radar signal (yHF(t)) to a baseband using the local oscillator signal (sLO(t)), and generating a digital signal (y[n]) based on the down-converted signal (y(t)); detecting first samples affected by interference in a portion of the digital signal (y[n]) corresponding to the scan chirp, and updating the start frequency (fSTART) for the regular chirps following the scan chirp based on the detected first samples; and detecting one or more radar targets based on the digital signal (y[n]), replacing the portion of the digital signal (y[n]) that does not correspond to the regular chirps with preset samples.

Description

TECHNICAL AREA The present disclosure relates to the field of radar sensors, in particular radar detection techniques that enable improved target detection in the presence of interference. BACKGROUND Radar sensors are found in numerous detection applications that measure the distances and speeds of objects. In the automotive sector, there is a growing demand for radar sensors, which can be used in advanced driver-assistance systems (ADAS). Examples of ADAS include adaptive cruise control (ACC) and radar cruise control. These systems can be used to automatically adjust a vehicle's speed to maintain a safe distance from other vehicles ahead. Blind spot monitors, which use radar sensors to detect other vehicles in a vehicle's blind spot, are another example of ADAS.Autonomous vehicles, in particular, can use numerous sensors, such as radar sensors, to detect and locate objects in their environment. Information about the position and speed of objects within the range of an autonomous vehicle is used to help ensure safe driving. Modern radar systems use highly integrated RF circuits that can integrate all the core functions of a radar transceiver's RF front end into a single package (single-chip transceiver). Such RF front ends typically include, among other things, a local RF oscillator (LO), power amplifiers (PA), low-noise amplifiers (LNA), and a mixer. Frequency-modulated continuous-wave (FMCW) radar systems use radar signals whose frequency is modulated by ramping up and down the signal frequency. Such radar signals are often referred to as "chirp signals" or simply "chirps." In the case of linear chirp signals, the term "LFM signals" is sometimes used, where LFM stands for "linear frequency modulation."A radar sensor typically emits sequences of chirps using one or more antennas, and the emitted signal is backscattered by one or more objects (called radar targets) within the radar sensor's field of view. The backscattered signals (radar echoes) are received and processed by the radar sensor. Detection of the radar targets is usually achieved using digital signal processing. As more and more cars are equipped with radar sensors, interference is becoming a concern. This means that the radar signal emitted by a first radar sensor (installed in a car) can interfere with the receiving antenna of a second radar sensor (installed in another car) and disrupt its operation. Publication WO 2016164472 A1 describes an example of an FMCW radar system. An example of a radar system that employs a technique to reduce interference-induced disturbances is described, for instance, in publication US 20190056476 A1. One of the problems underlying the invention can be seen as improving and extending existing concepts for suppressing interference-induced disturbances in radar systems. The aforementioned problem is solved by the methods according to claims 1, 6, and 11, as well as by the radar system according to claim 15. Various embodiments and further developments are the subject of the dependent claims. OVERVIEW The aforementioned problem is solved by the methods according to claims 1 and 7 and the system according to claim 10. Various embodiments and further developments are the subject of the pending claims. According to one embodiment, the method comprises generating the frequency-modulated local oscillator signal with a frame partitioned into N chirp positions (where N is a predetermined integer), wherein the frame has at least one scan chirp at one or more chirp positions, as well as regular chirps at the remaining chirp positions, wherein the regular chirps have a chirp bandwidth and a start frequency, and the at least one scan chirp has a scan bandwidth that is higher than the chirp bandwidth.The method further comprises transmitting an RF signal representing the frame, wherein the transmit power is reduced to zero or below the power of a regular chirp during the transmission of the at least one scan chirp; receiving an RF radar signal corresponding to the frame, down-mixing the received RF radar signal to the baseband using the local oscillator signal, and generating a digital signal based on the down-mixed signal; detecting noisy samples affected by interference in a portion of the digital signal corresponding to the scan chirp and updating the start frequency for the regular chirps following the scan chirp based on the detected noisy samples; and detecting one or more radar targets based on the digital signal, wherein the portion of the digital signal not corresponding to the regular chirps is replaced by preset samples. According to a further embodiment, the method comprises generating a frequency-modulated local oscillator signal with multiple frames of chirps and one or more scan chirps between the frames; generating an RF signal, wherein the transmit power of the RF signal is reduced or zero during the generation of the at least one scan chirp; receiving an RF radar signal, down-converting the received RF radar signal to a baseband using the local oscillator signal, and generating a digital signal based on the down-converted signal, wherein the digital signal is composed of multiple sequences, each sequence corresponding to either a chirp, a specific frame, or a scan chirp; repeatedly generating, for each sequence, metadata indicating whether the sequence in question is disturbed by interference; and repeatedly setting a start frequency for the frame chirps based on the metadata. Furthermore, this document describes the relevant radar systems and equipment. BRIEF DESCRIPTION OF THE DRAWINGS The invention can be better understood with reference to the following drawings and descriptions. The components in the figures are not necessarily to scale; instead, the emphasis has been placed on illustrating the principles of the invention. In the figures, identical reference numerals denote corresponding parts. Regarding the drawings: Fig. 1 is a sketch illustrating the operation of an FMCW radar system for distance and / or speed measurement. Fig. 2 contains two time-history diagrams illustrating the frequency modulation of the RF signal used in FMCW radar systems. Fig. 3 illustrates an example of how interference is introduced onto the receiver of a radar sensor. Fig. 4 is a time-history diagram illustrating a frame of chirps used for data acquisition in a radar sensor.Figure 5 illustrates, in a time-history diagram, a transmitted signal from a radar sensor and an interference signal from a jammer, with the frequency-over-time curves of these signals at least partially overlapping. Figure 6 illustrates an exemplary curve with a radar signal (after downmixing to the baseband) containing a radar echo from a target and interference shown in Figure 5. Figure 7 is a block diagram illustrating the basic structure of an FMCW radar system. Figure 8 is a circuit diagram illustrating an example of an analog RF front end of a radar sensor and an analog RF front end of a jammer. Figure 9 shows an example of a chirp frame with a preceding scan chirp. Figure 10 illustrates a concept for determining suitable sub-bands for transmitting the chirps of a chirp frame. Figures 11 and 2 are shown in Figure 2.Figure 12 illustrates the detection of noisy samples within a chirp frame and the replacement of regular chirps with scan chirps in a chirp frame. Figure 13 illustrates the continuous generation of chirps by inserting a sequence of scan chirps into the frame gap. Figure 14 illustrates a radar system with an MMIC and a separate controller chip in which radar data and metadata are processed independently. Figures 15 and 16 contain flowcharts illustrating examples of metadata processing. DETAILED DESCRIPTION Fig. 1 shows a simplified example of a conventional frequency-modulated continuous wave (FMCW) radar sensor 1. In this example, separate transmit (TX) and receive (RX) antennas 5 and 6, respectively, are used (bi-static or pseudo-monostatic radar configuration). However, it should be noted that a single antenna can be used, so that the receive and transmit antennas are physically the same (monostatic radar configuration). The transmit antenna 5 emits a (quasi-)continuous RF signal sHF(t), which is frequency-modulated, for example, by a sawtooth waveform. When the emitted signal sHF(t) is backscattered by an object T, which may be located in the radar channel within the radar's detection range, the backscattered signal yHF(t) is received by the receive antenna 6. The object T is commonly referred to as the "radar target".In a more general example, there may be more than one target within the field of view of a radar sensor, and an antenna array may be used instead of a single RX antenna. Similarly, an antenna array may be used instead of a single TX antenna. The use of multiple RX and TX antennas in a multi-channel radar system allows the measurement of the angle of arrival of a radar echo, commonly referred to as the direction of arrival (DoA). Measuring the DoA is important for many applications, and therefore most radar sensors use antenna arrays. To keep the diagrams simple, only one TX antenna and one RX antenna are shown in the figures. It is understood that the concepts described herein are readily transferable to radar sensors with antenna arrays. Fig. 2 illustrates the aforementioned frequency modulation of the signal sHF(t). As shown in the upper diagram of Fig. 2, the signal sHF(t) consists of a series of "chirps," i.e., pulsed sinusoidal waveforms with increasing (upward chirp) or decreasing (downward chirp) frequency. In this example, the instantaneous frequency fLO(t) of a chirp increases linearly from a starting frequency fSTART to a final frequency fSTOP within a defined time interval TCHIRP (see the lower diagram of Fig. 2). Such a chirp is also referred to as a linear frequency ramp. A linearly frequency-modulated (LFM) signal with three identical linear frequency ramps is shown in Fig. 2. However, it should be noted that the parameters fSTART, fSTOP, TCHIRP, as well as the pause between each frequency ramp, may vary depending on the actual implementation of the radar system 1 and may also change during operation of the radar system.In practice, the frequency change can be, for example, linear (linear chirp, linear frequency ramp), exponential (exponential chirp) or hyperbolic (hyperbolic chirp), although linear chirps are most commonly used. Figure 3 illustrates a simple example showing how the operation of a radar sensor can be disrupted by other radar sensors, referred to as jammers. Accordingly, Figure 3 shows a three-lane road with five vehicles A1, A2, A3, A4, and A5. In this example, vehicle A1 is considered the "own vehicle" and its radar sensor the "own radar sensor." The own radar sensor transmits a signal and detects an echo signal E1 backscattered from vehicle A5, which represents the radar target T to be detected by the own radar sensor. However, in addition to the desired echo signal E1, the own radar sensor also receives interfering signals D2, D3, and D4, which are transmitted by the onboard radar sensors (jammers) of vehicles A2, A3, and A4.These interfering signals interfere with the desired radar echo E1 and can negatively affect the detection of radar targets from the received radar signal (which contains the echo E1 as well as the interfering signals D2, D3 and D4). Figure 4 schematically illustrates an example of a frequency modulation scheme as commonly implemented in FMCW radar sensors. In the example shown, a sequence of sixteen upward chirps is transmitted for data acquisition. Such a sequence of a defined number of chirps is called a "chirp frame." It should be noted that in practice, a chirp frame usually contains many more chirps (e.g., 256 chirps), and the present example has been simplified for illustrative purposes only. A radar sensor transmits a frequency-modulated RF signal, as shown in Figure 4, and receives a corresponding RF radar echo signal. An exemplary signal processing method for evaluating the radar echoes will be discussed later with reference to Figure 12. At this point, it should be noted that one frame is used for acquiring radar data (digitized baseband radar signal).In particular, the data corresponding to a chirp frame can form the complete input data for a range-Doppler FFT, which makes it possible to determine distance and velocity information from the data corresponding to a single chirp frame. Depending on the characteristics of the interfering signals (see Fig. 3, signals D2, D3, and D4), the desired radar echoes can be affected in various ways. Figs. 5 and 6 illustrate, using an example, how a jammer can disrupt the received radar echoes if the interfering signals contain chirps that have different parameters, in particular a different frequency slope, than the radar echoes. Fig. 5 illustrates the frequency of a chirp over time (chirp duration 60 µs) emitted by the radar sensor itself in the example of Fig. 3. The starting frequency of the emitted signal sHF(t) is approximately 76250 MHz, and the ending frequency is approximately 76600 MHz. An interfering signal (e.g., the interfering signal D3 in the example of Fig. 3)3), generated by another radar sensor, contains an upward chirp that begins at approximately 76100 MHz and ends at 76580 MHz (chirp duration 30 µs), and a subsequent downward chirp that begins at the end frequency of the preceding upward chirp (76580 MHz) and ends at the start frequency of the next upward chirp (76100 MHz), with a chirp duration of 10 µs. The bandwidth BWBB of the baseband signal of the in-house radar sensor is indicated by dashed lines in Fig. 5. It is noted that when an IQ mixer (in-phase / quadrature mixer) is used to convert the received RF signal to baseband, the resulting digital baseband signal can be considered a complex-valued (analytical) signal which, depending on how the IQ mixer is implemented, contains either only positive or only negative frequency components (spectral lines). Fig. 6 shows an example of the waveform of the (preprocessed) baseband signal resulting from the received radar signal, which contains a desired radar echo and interference. It can be seen that the signal components exhibit significant levels of interference during those time intervals in which the frequency of the interference signal lies within the bandwidth BWBB of the radar sensor (see Fig. 5). In this example, the interference occurs three times during the 60 µs chirp duration, at approximately 7 µs, 28 µs, and 42 µs. As mentioned, the power of interference signals is typically higher than the power of radar echoes from actual targets. Accordingly, the power of the received RF signal increases when it is disturbed. Nevertheless, interference appears as relatively short bursts, the duration of which depends on the length of time the bandwidths of the signals overlap.Therefore, not all chirps of a frame (see Fig. 4) are typically affected by interference. Furthermore, in this example, interference signals and the transmitted signal of a particular radar sensor are uncorrelated, so the interference can be considered noise and increases the overall background noise. The baseband signal shown in Fig. 6 is processed as a digital signal that can be partitioned into frames corresponding to the chirp frame in the emitted RF radar signal. Each frame of the digital signal contains multiple digital samples, and a frame can again be partitioned into segments corresponding to the individual chirps of a frame in the emitted RF radar signal. As mentioned above, the power of the received RF signal (and thus the power of the resulting baseband signal) is higher when it contains interference, compared to a situation where it contains only echo signals from real radar targets.Therefore, samples affected by interference can be easily detected using a digital threshold detector. Before discussing some new concepts for attenuating / preventing interference, an example of a radar sensor, and in particular an RF front-end of a radar sensor, will be explained in more detail, which should aid in understanding the subsequent discussion. The examples discussed herein concern a radar system with an RF front-end that uses a mixer producing a real-valued baseband signal. It is understood that the concepts described herein can readily be transferred to radar systems with RF front-ends that use an IQ mixer producing a complex-valued (i.e., analytical) signal.Downmixing / demodulation using IQ mixers is a technique that is well known in the field of radar sensors and is therefore not explained separately here. Figure 7 is a block diagram illustrating an example structure of a radar sensor 1. Accordingly, at least one transmitting antenna 5 (TX antenna(s)) and at least one receiving antenna 6 (RX antenna(s)) are connected to an RF front end 10, which may be integrated in a semiconductor chip commonly referred to as a monolithic microwave integrated circuit (MMIC). The RF front end 10 may contain all the circuit components required for RF signal processing. Such circuit components may include, for example, a local oscillator (LO), RF power amplifiers, low-noise amplifiers (LNAs), directional couplers such as rat-race couplers and circulators, and mixers for downmixing (demodulating) RF signals (e.g., the signal yHF(t), see Figure 1) to the baseband (also referred to as the IF band). As mentioned, antenna arrays can be used instead of individual antennas.The illustrated example shows a bi-static (or pseudo-monostatic) radar system that has separate RX and TX antennas. In the case of a monostatic radar system, a single antenna or antenna array can be used for both receiving and transmitting electromagnetic (radar) signals. In this case, a directional coupler (e.g., a circulator) can be used to separate RF signals to be transmitted to the radar channel from RF signals received by the radar channel. In the case of an FMCW radar sensor, the RF signals radiated by the TX antenna 5 can be in a range between approximately 20 GHz (e.g., 24 GHz) and 81 GHz (e.g., about 77 GHz in automotive applications). As mentioned, the RF signal received by the RX antenna 6 contains the radar echoes, i.e., the signals backscattered from the radar target(s), and, if applicable, interference signals emitted by other radar sensors (interference sources). The received RF signal yHF(t) is downconverted to the baseband and further processed in the baseband using analog signal processing (see Fig. 8, baseband signal processing chain 20), which essentially involves filtering and amplifying the baseband signal and thus determines the bandwidth of the received signal (cf. Fig. 5, bandwidth B). The baseband signal is finally digitized using one or more analog-to-digital converters 30 and in the digital domain (see Fig.3 , digital signal processing chain, which is implemented, for example, in the digital signal processor 40). The overall system is controlled by a system controller 50, which is implemented, at least in part, using a processor running suitable firmware. The processor can be contained, for example, in a microcontroller, a digital signal processor, or the like. The digital signal processor 40 (DSP) can be part of the system controller 50 or separate from it. The signal processor and the system controller can be integrated in an application-specific integrated circuit (ASIC). The RF front end 10 and the analog baseband signal processing chain 20 (and optionally the ADC 30) can be integrated in a single MMIC. However, the components shown in Fig. 7 can be distributed across two or more integrated circuits. In particular, some parts of the digital signal processing can be performed in the MMIC.Fig. 8 shows an exemplary implementation of the RF front-end 10, which may be incorporated into the radar sensor shown in Fig. 7. It should be noted that Fig. 8 is a simplified circuit diagram illustrating the basic structure of an RF front-end. Actual implementations, which can vary considerably depending on the application, may of course be more complex. In particular, many practical implementations incorporate multiple receive and transmit channels, whereas in the illustrated example, only one receive channel (formed, among other things, by the LNA 103, the mixer 104, and the baseband processing chain 20) and one transmit channel (formed, among other things, by the power amplifier 102) are shown for the sake of simplicity. The RF front-end 10 incorporates a local oscillator (LO) 101, which generates an RF signal sLO(t) that, as explained above with reference to Fig. 2 and Fig. 4, may be frequency-modulated. The signal sLO(t) is also referred to as the LO signal.In radar applications, the LO signal is usually in the SHF (super high frequency) or EHF (extremely high frequency) band, e.g. in automotive applications between 76 GHz and 81 GHz. The LO signal sLO(t) is processed in the transmit signal path (transmit channel) and in the receive signal path (receive channel). The transmit signal sHF(t), radiated by the TX antenna 5, is generated by amplifying the (frequency-modulated) LO signal sLO(t), for example, using an RF power amplifier 102. The output signal of the amplifier 102 is coupled to the TX antenna 5, for example, via strip lines, a coupler, a matching network, etc. (not shown in Fig. 9). The transmit signal sHF(t) is also referred to as the outgoing radar signal. The received RF signal yHF,T(t), supplied by the RX antenna 6, is fed to a mixer 104. In the present example, the received RF signal yHF(t) (i.e. the antenna signal) is pre-amplified by an RF amplifier 103 (e.g. a low-noise amplifier, LNA, with gain g) so that the mixer receives the amplified signal g·yHF(t) at its RF input.Mixer 104 continues to receive the frequency-modulated LO signal sLO(t) at its reference input and is configured to downmix (demodulate) the amplified RF signal g·yHF(t) to the baseband. The resulting baseband signal at the mixer output is designated yBB(t). The baseband signal yBB(t) is further processed by the analog baseband signal processing chain 20 (see also Fig. 8), which essentially comprises one or more filters (e.g., a bandpass filter or highpass and lowpass filters) to remove unwanted sidebands and frame rates, as well as one or more amplifiers. The analog output signal of the baseband signal processing chain 20 is designated y(t) and can be fed to an analog-to-digital converter (ADC) 30 (see also Fig. 8). The digital signal y[n] output by the ADC 30 is referred to as the digital radar signal, which contains the digital radar data. It is understood that the digital radar signal is partitioned into frames corresponding to the chirp frames of the LO signal sLO(t), and that each frame of the digital radar signal can be subdivided into multiple segments corresponding to the multiple chirps in the respective chirp frame. Acquisition (i.e., a measurement process) requires the acquisition of a frame of digital radar data, with the data acquisition being repeated at a defined (frame) repetition rate. The digital radar signal (e.g., frame by frame) can be fed to a processor, such as a digital signal processor 40, which is programmed to further process the digital radar signal, e.g., by applying algorithms collectively known as range / Doppler processing. The implementation of the circuit components shown in Fig. 8 is well-known in the context of a radar sensor and is therefore not described in detail. Fig. 8 also shows how a desired radar echo and a radar signal transmitted by another radar sensor interfere. Fig. 8 further shows the radar front end 10' of another radar sensor, showing only the local oscillator 101', the transmit channel (with the amplifier 102'), and the transmit antenna 5' for simplicity. The other radar sensor emits a signal sHF'(t). The resulting RF signal arriving at the receiving antenna 6 of the first radar sensor is called the RF jam signal yHF,I(t). The receiving antenna 6 of the first radar sensor receives the RF jam signal yHF,I(t) along with the desired RF echo signal yHF,T(t) caused by the radar target T from which the signal sHF(t) of the first radar sensor is backscattered. Both the radar echo yHF,T(t) and the interference signal yHF,I(t) are received by the antenna 6 and superimpose at the RF input of the mixer 104 (yHF(t) = yHF,T(t) + yHF,I(t)). Figure 8 shows that the interference signal component yHF,I(t) of the received signal yHF(t) is down-converted to the baseband in the same way as the radar echoes yHF,T(t) contained in the received signal yHF(t). Accordingly, if the frequency difference between the instantaneous frequency fLO of the transmitted signal sHF(t) and the instantaneous frequency of the received interference signal yHF,I(t) lies within the bandwidth BWBB of the baseband signal processing chain 20, the interference is also present in the digital radar signal y[n]. Several concepts exist for mitigating interference in radar sensors. Some of these concepts can be combined. According to some approaches, frames or individual segments (corresponding to specific chirps) that are affected by interference are simply detected and discarded. Other approaches aim to cancel out the signal components caused by interference and restore the original, undisturbed signal in the digital domain. The concept described below aims to avoid interference by modifying chirp parameters (especially the start frequency of a chirp parameter) before, at the beginning of, or within a chirp frame. For the following discussion, it is assumed that the local oscillator of a radar sensor can be tuned over a relatively wide frequency band. For example, the local oscillator allows the LO frequency fLO of the LO signal sLO(t) to be tuned from 76 GHz to 81 GHz. That is, in this example, the maximum bandwidth BWMAX of the radar sensor is 5 GHz. However, during normal operation, the bandwidth BW of the chirps used for radar data acquisition can be significantly smaller, for example, 1–3 GHz. This means that the "position" of the chirp bandwidth BW can be shifted within the maximum bandwidth BWMAX without changing anything in the subsequent signal processing used for target detection. It is noted that the linearity of the frequency modulation is important for the quality of radar measurements. Although ordinary implementations of local oscillators (usually a voltage-controlled oscillator coupled in a phase-locked loop) can be tuned over a relatively large bandwidth of 5 GHz, the achievable linearity of a 5 GHz frequency ramp (referred to as a broadband chirp) would be insufficient for normal radar measurements. By tuning the local oscillator only through the relatively small chirp bandwidth of, for example, 1 GHz, significantly better linearity can be achieved. Figure 9 illustrates an exemplary implementation of a concept that can be described as frequency hopping (although this term has a slightly different meaning in the field of wireless communication). Figure 9 contains a frequency-versus-time diagram illustrating the instantaneous frequency fLO(t) of the LO signal sLO(t), which, in the example of Figure 8, can be generated, for instance, by the local oscillator 101. Accordingly, the LO signal sLO(t) contains several chirp frames, each composed of N chirps (where N is a positive integer greater than one, e.g., N = 256), with the bandwidth BW of the N chirps in a chirp frame being less than the maximum bandwidth BWMAX. The start frequency of each chirp is denoted by fSTART, and the stop frequency fSTOP is therefore fSTART + BW for an upward chirp. During operation, an amplified version of the chirp frame is transmitted as an RF radar signal sHF(t).Before the first chirp of a chirp frame is generated, a broadband chirp is produced, which has a higher bandwidth than the regular chirps of a chirp frame. In the illustrated example, the broadband chirp has a bandwidth equal to the maximum bandwidth BWMAX. The broadband chirp is a "passive" chirp, meaning that the transmitter power is significantly reduced or even zero while the local oscillator is generating the broadband chirp. For example, the power amplifier 102 (see Fig. 8) can be deactivated during a broadband chirp. A passive broadband chirp allows the full bandwidth fMAX - fMIN = BWMAX to be scanned for potential interference. Therefore, a passive broadband chirp is also called a "scan chirp". Since a scan chirp is passive, the received RF signal yHF(t) does not contain any radar echoes; however, if an interference source is present near the radar sensor, the received RF signal will nevertheless contain the interference signal yHF,I(t) caused by the interference source (see also Fig. 8). In the radar sensor's receive channel, the interference signal yHF,I(t) is demodulated using the broadband chirp, and—if an interference source is present—the interference caused by one or more interference sources becomes apparent in the resulting (digital) baseband signal y[n], for example, as shown in Fig. 6. Since the scan chirp is a broadband chirp, a relatively large bandwidth (or the full BWMAX bandwidth of the radar sensor) is "scanned" for potential interference.The signal segment of the digital radar signal y[n] corresponding to the scan chirp can be analyzed to detect a suitable frequency band with chirp bandwidth BW within the full frequency range from fMIN to fMAX. In this context, a suitable frequency band means a frequency band with bandwidth BW that contains no or very little interference. The start frequency fSTART of the (active) chirps of a chirp frame used for radar data acquisition can then be modified so that the frequency range from fSTART to fSTOP = fSTART + BW contains no or as little interference as possible. An efficient approach for detecting suitable sub-bands for transmitting a chirp frame is described below.In radar systems with multiple input (receive) channels and multiple output (transmit) channels—so-called MIMO systems—only a single input / receive channel can be active while generating a scan chirp, in order to reduce the radar system's power consumption. Since scan chirps are only used to detect jammers (and not actual radar targets), a single active input channel is sufficient. The other input / receive channels are only used during the time intervals when the regular chirps of a chirp frame are being transmitted. Fig. 10 shows in the upper diagram a scan chirp covering the full BWMAX bandwidth, as well as the corresponding signal segment of the digital radar signal y[n]. As can be seen in the middle diagram of Fig. 10, the digital radar signal y[n] contains interference caused by one or more interfering sources. It is understood that the digital radar signal y[n] is a discrete-time signal composed of several digital samples belonging to discrete time points. The lower diagram of Fig. 10 shows a signal P[n] representing the instantaneous power of the digital radar signal. In one example, the signal P[n] can simply represent the envelope of the digital radar signal P[n]. For the concept described herein, it is not important how the signal P[n] is actually defined and calculated. The purpose of Fig.The objective of Section 10 is to demonstrate that samples of the digital radar signal y[n] that are perturbed by interference can be easily detected using a threshold detection technique. That is, any sample whose power exceeds a defined threshold is detected as perturbed by interference. It is understood that this is equivalent to detecting samples whose magnitude exceeds a corresponding threshold TH. As stated above, the concepts described herein can be readily transferred to radar systems that use IQ mixers for down-mixing the received RF radar signal to the baseband. In this case, the digital radar signal can be considered a complex-valued signal for which the magnitude can be calculated as the square root of the sum of the squares of the real and imaginary parts.To avoid the need to calculate the square root, the squared amount (representing a signal power) can be compared to a corresponding threshold value. In the example shown in Fig. 10, top diagram, it can be seen that each time point t1, t2, t3, etc., is assigned to a corresponding frequency f1, f2, f3, etc. Since the scan chirp is a linear frequency ramp, each sample of the digital radar signal is assigned to a corresponding (discrete) frequency; and in the case where a particular sample is detected as being disturbed by interference, the corresponding discrete frequency can be considered "blocked by interference." Conversely, all other samples that are not detected as being disturbed by interference correspond to corresponding discrete frequencies that are "free," i.e., not blocked by interference. As can be seen in Fig. 10, the result of this detection, although undisturbed samples are detected in the time domain (e.g., by comparing the magnitude of each sample to a threshold TH sample by sample), can be directly assigned to free and blocked frequencies. In the example of Fig. 10, the time interval from t1 to t2, which contains no samples detected as noisy, can be directly assigned to the frequency interval from f1 to f2. The sample-by-sample detection of undisturbed samples in the time domain directly yields free frequency intervals. In the present example, the start frequency fSTART can be set to a frequency equal to or slightly above f1 to shift the chirps of the subsequently transmitted chirp frame into a free sub-band. It should be noted here that no calculations are required to determine the free sub-bands. The sample rate, and therefore the sample interval Δt, of ADC 30 (see Fig. 10)8) is a known system parameter, and due to the use of linear frequency ramps, the time interval Δt is proportional to a frequency increment Δf, which is therefore also a known system parameter and depends on the steepness of the frequency ramp. In a segment of a digital radar signal corresponding to a scan chirp, the first sample can be assigned to the start frequency, the second sample to the frequency fSTART+Δf, the third sample to the frequency fSTART+2Δf, and so on. According to some embodiments described herein, the start frequency fSTART of a chirp (and thus the sub-band in which the chirp is generated) can be changed within a chirp frame. For this purpose, a regular chirp within a chirp frame is replaced by a scan chirp. The replacement of regular chirps by a scan chirp within a chirp frame can generally occur at any position of a chirp within the chirp frame. For example, the 10th regular chirp, the 86th regular chirp, or the 156th regular chirp can be replaced. Furthermore, not only one regular chirp but more than one regular chirp within the chirp frame can be replaced. The decision as to whether a regular chirp is replaced by a scan chirp, and which chirp is replaced, can depend on the detected disturbance. Fig. 11 illustrates a situation in which the operation of a radar sensor is disrupted by two jammers. The diagram in Fig.Figure 11 illustrates a portion of a chirp frame transmitted during radar data acquisition. The solid line represents the frequency of the radar sensor's local area (LO) signal, sLO(t). The dashed and dotted lines represent the frequencies of interfering signals injected into the radar sensor's receive channel. The shaded areas mark the regions (i.e., time intervals) where the instantaneous frequencies of the interfering signals lie within a certain corridor around the frequency fLO of the radar sensor's LO signal, sLO(t), and thus where interference can occur. The width of this corridor depends on the bandwidth BWBB of the baseband signal (see also Figure 5 or Figure 10). Within a chirp frame, which has N chirp positions (i.e., N time slots in which a chirp can be generated), noisy samples can be repeatedly detected (e.g., sample by sample), with each noisy sample being assigned a corresponding "blocked" frequency, while the remaining frequencies can be considered "free" (without interference). In Fig. 12, the aforementioned chirp positions are labeled "Chirp 1," "Chirp 2," "Chirp 3," etc. The chirp positions define the temporal positions (i.e., time slots) of the chirp within a chirp frame. The time slots of a frame have a defined length (chirp length TCHIRP) and are the same across a chirp frame. Within a chirp frame, a controller (see Fig. 7) can decide, based on the samples detected as noisy, to insert a scan chirp into the current chirp frame.Accordingly, a scan chirp is generated at one or more chirp positions, as shown in the example in Fig. 12. This means that the regular chirps scheduled for these chirp positions are replaced by the scan chirp. As mentioned above, the transmit power (compared to the regular transmit power) is reduced or zero while the scan chirp is being generated. Furthermore, it is important to maintain the timing of the chirp positions / time slots. That is, at least one scan chirp is emitted at one or more consecutive chirp positions, whereby a scan chirp may contain a modulation pause during which the LO signal is not modulated (see Fig. 12, pause at chirp position "Chirp 7"). The modulation pause can be used for the calculations and signal processing required to determine a free sub-band. Based on the samples (during the scan chirp and during the preceding regular chirps) detected as interference, and the corresponding frequencies, a clear sub-band (or a sub-band with only a few interferences) can be selected, and the start frequency fSTART can be set for the subsequent chirps of the chirp frame, as shown in Fig. 12. The digital radar data (i.e., the samples of the digital radar signal) corresponding to the chirp positions "Chirp 6" and "Chirp 7" occupied by the scan chirp can be replaced by preset values, e.g., zero (zero padding). Zero padding those segments of a frame of the digital radar signal corresponding to the chirp positions occupied by the scan chirp (and, if applicable, the modulation pause) may slightly reduce the quality of target detection.However, the effect of null padding becomes relatively small (or even negligible) if, in a frame that typically contains a relatively large number (e.g., N=256 or N=512) of chirp positions, only one or a few regular chirps are replaced by scan chirps. The resulting frame can still be processed for target detection using range-Doppler signal processing techniques. In the examples discussed so far, only a single chirp frame has been considered. However, it is understood that a radar sensor transmits a multitude of successive chirp frames over multiple transmission channels (in multi-channel systems) during normal operation. In the example of Fig. 13, a frame interval is used to generate passive scan chirps to monitor for interference regularly and continuously, and to use the interference information to periodically adjust the start frequency fSTART of the (active) regular chirps. In the example of Fig. 13, the scan chirps are passive chirps (transmitter power low or zero), but they do not need to have a bandwidth higher than the regular chirp bandwidth BW. Nevertheless, it can be useful to generate broadband chirps as scan chirps. In some embodiments, the frame interval can also be partitioned into chirp positions / time slots that have the same length TCHIRP as the chirp positions / time slots used for radar data acquisition (see Fig. 12). The information about which samples are detected as being disturbed is hereafter referred to as metadata (because it represents data about digital radar data). In one example, the metadata contains one bit (disturbed / undisturbed) per sample of the digital radar data provided by the ADC 30 (see Fig. 8). It is noted that in this case, as discussed with reference to Fig. 10, each bit of the metadata can be directly assigned to a discrete frequency. The metadata can be generated and analyzed continuously (repeatedly for each chirp, sample, or group of samples).Based on the metadata, the start frequency fSTART of the regular chirps in the chirp frame can be updated. This update can occur before a new data acquisition begins and a new chirp frame starts, or even within a chirp frame. It is understood that the implementations and concepts described herein can be combined. Fig. 14 shows a block diagram of a radar system in which parts of the system controller 50 and the signal processing unit 40 (see Fig. 7) are integrated on a separate chip. As mentioned, the controller 50 and the digital signal processing unit 40 are considered functional units that may have several sub-units, which may be distributed across two or more chips. The MMIC 1 contains one or more receive channels RX0, RX1 and one or more transmit channels TX0, TX1, which may be configured similarly to the example in Fig. 8. The local oscillator 101 generates the (frequency-modulated) LO signal sLO(t), which can be amplified and transmitted as RF radar signals via the transmit channels. Furthermore, the LO signal sLO(t) is used to demodulate the received RF signals, as discussed with reference to Fig. 8. The resulting baseband signals are digitized using one or more analog-to-digital converters (see Fig. 14, ADC 30).The frequency modulation provided by the local oscillator LO 101 is controlled by a ramp generator 111, which is essentially a circuit that controls the chirp parameters such as start frequency fSTART, stop frequency fSTOP, chirp duration TCHIRP, modulation pause, etc. (see also Fig. 2). The functional block / processing unit / control circuit labeled "insert metadata" is configured to generate the metadata that, as discussed above (e.g., using a threshold TH, see Fig. 10), indicates which of the individual samples of the digital radar signal is affected by interference, and to mix the metadata and the digital radar data to produce a data stream that is sent to the controller chip 2 via a high-speed serial communication link (see Fig. 14, serial interfaces 31 and 51).It is noted that the control circuit that inserts the metadata into the radar data stream provided by the ADC 30 can be configured such that the data stream transmitted over the communication link contains metadata only in the time intervals between chirp frames (frame gaps) when only (passive) scan chirps are generated. This avoids overloading the memory (see Fig. 14, memory 53) of the MCU 2 shown in Fig. 14. In the controller chip 2, the data stream received via the serial high-speed communication link is separated into the digital radar data and the metadata. The metadata can be buffered in a memory 54. The digital radar data can be processed in any known way (see Fig. 13, block 52 and memory 53) to detect radar targets, e.g., using range-Doppler processing, CFAR algorithms, etc. (CFAR = Constant False Alarm Rate). The metadata stored in memory 54 can be processed continuously (see Fig. 15, DSP 55) to determine a suitable free or optimal sub-band (for example, defined by a start frequency fSTART and a chirp bandwidth BW) for the regular chirps emitted during the chirp frame. The system structure of Fig.14 enables the processing of metadata independently and simultaneously with range / Doppler processing, and the processing of the metadata does not burden the range / Doppler processor (see Fig. 14, processor unit 52). Because the metadata is separated from the digital radar data (i.e., the digital radar signal y[n] provided by the ADC 30) before being stored / buffered in memory 54, the behavior of the range / Doppler processor 52 cannot be affected by the metadata generated during the time interval between the chirp frames (frame interval); such a concept makes it possible to use an associated processing unit 55, which can operate completely independently of the range / Doppler processor 52, to analyze the metadata. The system structure shown in Fig. 14 enables feedback from the processor 55, which regularly determines an updated start frequency fSTART, to the ramp generator 111 in the MMIC 1. This feedback of the updated start frequency and, if applicable, other ramp parameters is indicated by the dashed lines in Fig. 14. This information can be exchanged via an associated communication channel, such as a serial peripheral interface (SPI), direct line connections connected to high-speed input / output pins, or a common serial interface between the MMIC and the MCU, which is used for control, diagnostics, and updating of the start frequency fSTART (and other chirp parameters) used by the ramp generator 111. The functional separation between metadata processing (processor 55) and target detection (range Doppler processing 52) allows metadata processing to be implemented using firmware that is programmed once into the controller chip 2 by the chip manufacturer and cannot be modified by the customer, whereas the software implementing target detection is under the customer's control in many applications and can therefore be modified by the customer. During radar system operation, metadata processing (processor 55) and target detection (range Doppler processing) operate simultaneously, and any change / update to the start frequency of the regular chirps in a chirp frame does not require corresponding changes in target detection. Figure 15 illustrates an example of metadata processing by processor 55 in the system of Figure 14. Accordingly, the metadata received by MMIC 1 via the high-speed serial communication link is analyzed (see Figure 15, step S101) to determine which frequencies of a given regular chirp within a chirp frame are affected by interference. As discussed above with reference to Figure 10, each sample (taken at a specific point in time) can be directly assigned to a specific discrete frequency. Based on the metadata analysis, the following situations can be distinguished.In the present example, the metadata analysis involves dividing the current sub-band (frequency range from fSTART to fSTOP) of the regular chirps into an upper and a lower part, and determining whether interference occurs in both the upper and lower parts of the current sub-band (see Fig. 15, case distinction S102), or only in the lower part (see Fig. 15, case distinction S103), or only in the upper part (see Fig. 15, case distinction S104). If both parts of the current sub-band contain interference, it can be checked whether a free sub-band exists at higher or lower frequencies (i.e., frequency ranges above or below the current sub-band, see Fig. 15, case distinction S105). This information may be available from a previous full-band scan using scan chirps, as discussed above, e.g., with reference to Fig. 13.If a free sub-band is found at higher frequencies (i.e., frequency range above the current sub-band), the start frequency fSTART (and therefore the position of the current sub-band) is shifted to a higher frequency (see Fig. 15, step 106) so that a free (or less noisy) sub-band is used for the subsequent chirps. If a free sub-band is found at lower frequencies (i.e., frequency range below the current sub-band), the start frequency fSTART is shifted to a lower frequency (see Fig. 15, step S107). In a situation where metadata analysis reveals that only the lower part of the current sub-band contains noisy frequencies, the start frequency fSTART for subsequent chirps of the current chirp frame can be shifted to higher frequencies (see Fig. 15, step S106). For example, the new / updated start frequency fSTART can be set directly above the highest frequency identified as noisy. Alternatively, the new / updated start frequency fSTART can be set to a frequency in the upper part of the current sub-band. In a situation where metadata analysis reveals that only the upper part of the current sub-band contains noisy frequencies, the start frequency fSTART for subsequent chirps of the current chirp frame can be shifted to lower frequencies (see Fig. 15, step S107).For example, the new / updated start frequency fSTART can be set so that the corresponding stop frequency fSTOP = fSTART + BW is directly above the lowest frequency identified as noisy. Alternatively, the new / updated start frequency fSTART can be set so that the corresponding stop frequency fSTOP is in the lower part of the current sub-band. It is understood that the decisions regarding whether and how to shift the start frequency fSTART for subsequent chirps can be implemented in many different ways. Furthermore, it is noted that the purpose of Fig. 15 is not to present a complete algorithm for updating the start frequencies of regular chirps within a chirp frame based on metadata analysis. Rather, Fig. 15 is intended as a guideline to enable a person skilled in the art to implement various solutions for shifting the current sub-band of regular chirps.The process steps / activities shown in the flowchart of Fig. 15 can be implemented in various different ways, depending on the actual application. Figure 16 illustrates, by means of a flowchart, another example of metadata processing by processor 55 in the system of Figure 14. It should be noted that the example in Figure 16 is not an alternative to the preceding example in Figure 15, as both examples can be combined. Figure 15 concerns the processing of metadata relating to regular chirps of a chirp frame, whereas Figure 16 concerns the processing of metadata relating to scan chirps generated in the frame gap (see Figure 13), which may have a higher bandwidth than regular chirps of a chirp frame. According to Figure 16, the metadata of the scan chirps is analyzed (see Figure 16, step S201). This analysis (see Figure 16, step S201) may include checking which frequencies are disturbed, as discussed above with reference to Figure 10.As explained above, each sample corresponds to a specific time and a specific discrete frequency. Accordingly, any noisy frequency can be marked (see Fig. 16, step S202), and any coherent group of frequencies not marked as noisy can be considered free sub-bands (Fig. 16, step S203). This makes it possible to select a suitable start frequency fSTART within the free sub-bands for the upcoming regular chirps of the next chirp frame. The following is a summary of some exemplary embodiments described above. However, it should be emphasized that the following numbering, which contains 20 items, is not exhaustive but rather an overview of exemplary embodiments included in the description above. Example 1: Method comprising: Receiving an RF radar signal (see Fig. 8, signal yHF(t)); Down-converting the received RF radar signal to a baseband using a frequency-modulated local oscillator signal (see Fig. 8, signal sLO(t)) with a scan chirp having a bandwidth (see Fig. 9, bandwidth BW) higher than the bandwidth of a regular chirp; Generating a digital baseband signal (see Fig. 8, y[n]) based on the down-converted RF radar signal, wherein the digital baseband signal contains a sequence of samples belonging to the scan chirp; Identifying noisy samples, which are affected by interference (see Fig. 10), in the sequence of samples; Selecting a sub-band that has the regular chirp bandwidth, based on the position of the disturbed samples within the sequence of samples, to send chirps of a chirp frame used for measurement data acquisition (see Fig. 9 and Fig. 12). Example 2: Method according to Example 1, wherein the identification of impaired samples includes: checking the sequence of samples sample by sample to identify the samples representing an RF radar signal power above a threshold as impaired samples. Example 3: Method according to Example 1 or 2, wherein the positions of the perturbed samples correspond to frequency values ​​within the radar bandwidth, and wherein the sub-band is selected such that it contains no or as few frequency values ​​as possible that correspond to positions of perturbed samples. Example 4: Method according to one of Examples 1 to 3, wherein no or only negligible RF power is transmitted, while the received RF radar signal is down-converted to a baseband using the frequency-modulated local oscillator signal with the scan chirp. Example 5: A method according to any one of Examples 1 to 4, further comprising: frequency modulating the local oscillator signal so that it contains a sequence of chirps in the selected sub-band; transmitting an RF signal generated based on the local oscillator signal and containing the sequence of chirps in the selected sub-band; down-mixing the received RF radar signal to the baseband using the frequency-modulated local oscillator signal containing the sequence of chirps in the selected sub-band and generating the digital baseband signal; detecting one or more radar targets based on the resulting digital baseband signal. Example 6: Method comprising: generating a frequency-modulated local oscillator signal (see Fig. 8, signal sLO(t)) with a frame partitioned into N chirp positions, where N is a predetermined integer; wherein the frame contains at least one scan chirp at one or more chirp positions and regular chirps at the remaining chirp positions, the regular chirps having a chirp bandwidth and a start frequency, and the at least one scan chirp having a scan bandwidth higher than the chirp bandwidth (see Fig. 12); transmitting an RF signal representing the frame, wherein the transmit power is zero or reduced to below the power of a regular chirp during the transmission of the at least one scan chirp; receiving an RF radar signal (see Fig.8 , signal yHF(t)), corresponding to the frame, downmixing the received RF radar signal to a baseband using the local oscillator signal, and generating a digital signal (see Fig. 8 , signal y[n]) based on the downmixed signal; detecting noisy samples affected by interference in a portion of the digital signal corresponding to the scan chirp, and updating the start frequency for the regular chirps following the scan chirp based on the detected noisy samples; and detecting one or more radar targets based on the digital signal, with the portion of the digital signal not corresponding to the regular chirps being replaced by preset samples. Example 7: Procedure according to Example 6, wherein the frame contains a regular chirp at the first chirp position. Example 8: Method according to Example 6 or 7, wherein a scan chirp contains a frequency ramp extending over two or more chirp positions. Example 9: Method according to one of Examples 6 to 8, wherein a scan chirp contains a frequency ramp followed by a modulation pause. Example 10: Method according to one of Examples 6 to 9, wherein the part of the digital signal that does not correspond to the regular chirps is replaced by zero padding. Example 11: A method according to any one of Examples 6 to 9, wherein the digital signal contains N sub-sequences corresponding to the N chirp positions of the frame, each sub-sequence containing M samples; and wherein detecting the one or more radar targets includes: computing an N×M range Doppler map using M range Fourier transforms and N Doppler Fourier transforms; and detecting the one or more radar targets based on the N×M range Doppler map.Example 12: Method comprising: generating a frequency-modulated local oscillator signal (see Fig. 9, signal sLO(t)) with a plurality of frames of chirps and one or more scan chirps between the frames; generating an RF signal wherein the transmit power of the RF signal is reduced or zero during the generation of the at least one scan chirp; receiving an RF radar signal (see Fig. 8, signal yHF(t)), down-converting the received RF radar signal to a baseband using the local oscillator signal, and generating a digital signal based on the down-converted signal (see Fig.8 , signal y[n]), wherein the digital signal is composed of several sequences, each sequence corresponding either to a chirp of a particular frame or to a scan chirp; repeated generation of metadata indicating whether the sequence in question is disturbed by noise for each sequence; and repeated determination of a start frequency for the chirps of the frames based on the metadata (see Fig. 13 and Fig. 14 ). Example 13: Method according to Example 12, wherein the metadata for each chirp indicates which part(s) of a frequency band of the chirp in question is / are disturbed by interference. Example 14: Method according to Example 12 or 13, wherein the metadata allows inferences to be made about sample positions within each sequence of samples that are disturbed by interference. Example 15: Radar system comprising: a local oscillator configured to generate a frequency-modulated local oscillator signal (see Fig. 14, LO 101 and Fig. 8, signal sLO(t)) with multiple frames of chirps and one or more scan chirps; at least one transmit channel (see Fig. 14, channels TX01, TX02) configured to generate an RF signal (see Fig. 8, signal sHF(t)) based on the local oscillator signal, wherein the transmit power of the RF signal is reduced to zero or below the power of a normal chirp during the generation of the at least one scan chirp; at least one receive channel (see Fig.14 , channels RX01, RX02), configured to receive an RF radar signal, to downmix the received RF radar signal to a baseband using the local oscillator signal, and to generate a digital signal based on the downmixed signal, wherein the digital signal is composed of several sequences, each sequence corresponding either to a chirp of a particular frame or to a scan chirp; a first processing unit (see Fig. 14, functional block “inserting metadata”) configured to repeatedly generate metadata for each sequence indicating whether the sequence in question is disturbed by interference; and a second processing unit (see Fig. 14, metadata processor 55) configured to repeatedly determine a start frequency for the chirps of the frames based on the metadata. Example 16: Radar system according to Example 15, further comprising: a ramp generator (see Fig. 14, ramp generator 111) coupled to the local oscillator and configured to control the local oscillator according to one or more ramp parameters including the start frequency, in order to cause the local oscillator to generate the frequency-modulated local oscillator signal with the multiple frames of chirps and the one or more scan chirps. Example 17: Radar system according to Example 16, wherein the second processing unit is coupled to the ramp generator (see Fig. 14, connection indicated by dashed line) and is configured to update the start frequency used by the ramp generator. Example 18: Radar system according to one of Examples 15 to 17, wherein the local oscillator, the ramp generator, the at least one receive channel (RX01, RX02) and the first processing unit are integrated in an MMIC, wherein the second processing unit is integrated in a further integrated circuit (see Fig. 14, MCU 2), and wherein the MMIC and the further integrated circuit are connected by a communication link for sending the digital signal and the corresponding metadata from the MMIC to the further integrated circuit (see Fig. 14, fast serial interfaces 31 and 51). Example 19: Radar system according to Example 18, further comprising: a further connection between the MMIC and the further integrated circuit (see Fig. 14, connection indicated by a dashed line), wherein the second processing unit is configured to update the start frequency used by the ramp generator via the further connection. Example 20: Radar system according to one of Examples 15 to 19, further comprising: a third processing unit (see Fig. 14, range / Doppler processing) configured to detect one or more radar targets based on the digital signal, wherein the third processing unit operates independently of the second processing unit, and wherein the second processing unit is configured to operate independently of and simultaneously with the third processing unit. Example 21: Radar system comprising a local oscillator configured to generate a frequency-modulated local oscillator signal with a scan chirp having a higher bandwidth than a regular chirp, and further comprising at least one receive channel configured to receive an RF radar signal, to downmix the received RF radar signal to a baseband using the frequency-modulated local oscillator signal with the scan chirp, and to generate a digital baseband signal based on the downmixed RF radar signal, wherein the digital baseband signal contains a sequence of samples belonging to the scan chirp.The radar system further includes a processing unit designed to receive the sequence of samples belonging to the scan chirp, to identify disturbed samples in the sequence of samples that are affected by interference, and to select a sub-band, which has the bandwidth of a regular chirp, for sending chirps of a chirp frame used for measurement data acquisition, based on the position of the disturbed samples within the sequence of samples. Example 22: Radar system comprising a local oscillator configured to generate a frequency-modulated local oscillator signal with a frame partitioned into N chirp positions (where N is a predetermined integer), wherein the frame has at least one scan chirp at one or more chirp positions and regular chirps at the remaining chirp positions, the regular chirps having a chirp bandwidth and a start frequency, and the at least one scan chirp having a scan bandwidth greater than the chirp bandwidth. The radar system further comprises at least one transmit channel configured to send an RF signal representing the frame, wherein the transmitter power is reduced to zero or below the power of a regular chirp during the transmission of the at least one scan chirp.Furthermore, the radar system has at least one receive channel configured to receive an RF radar signal corresponding to the frame, to downmix the received RF radar signal to the baseband using the local oscillator signal, and to generate a digital signal based on the downmixed signal. A processing unit is configured to detect interference-affected, noisy samples in a portion of the digital signal corresponding to the scan chirp and to update the start frequency for the regular chirps following the scan chirp based on the detected interference chirps. The processing unit is further configured to detect one or more radar targets based on the digital signal, with the portion of the digital signal not corresponding to the regular chirps being replaced by preset samples. It is understood that a (signal) processing unit can be any unit (including hardware or software, or a combination of hardware and software) capable and configured to perform the desired procedures and functions described herein. In various embodiments, a processing unit can be implemented using electronic circuits configured to process digital and / or analog signals to provide the functions necessary to implement the procedures described herein. The electronic circuits may, for example, include digital processors and memory for storing instructions which, when executed by the processor, cause the electronic circuit to perform the desired functions.The electronic circuitry of a processing unit may also include peripheral circuitry that allows the processor to communicate and interact with other circuitry. In addition to, or as an alternative to, a software-executing processor, the electronic circuitry may include hard-wired logic circuitry that does not depend on software. The term "software" also includes firmware, which is typically programmed once into an integrated circuit. Other types of software may be application-specific and modifiable by a user. Processing units are sometimes also referred to as controllers, controller circuits, or controller units; or controllers, controller circuits, or controller units may contain one or more processing circuits. Even though various embodiments have been described with reference to one or more specific implementations, variations and / or modifications may be made to the examples presented without departing from the concept and scope of the appended claims. With particular regard to the various functions performed by the components or structures (units, assemblies, devices, circuits, systems, etc.) described above, the terms (including any reference to a "means") used to describe such components shall, unless otherwise specified, correspond to any component or structure that performs the specified function of the described component (e.g., that is functionally equivalent), even if it is not structurally equivalent to the disclosed structure that performs the function in the exemplary implementations of the invention presented herein.

Claims

A method comprising: generating a frequency-modulated local oscillator signal (sLO(t)) with a frame partitioned into N chirp positions, where N is a predetermined integer; wherein the frame contains a scan chirp at one or more chirp positions and regular chirps at the remaining chirp positions, the regular chirps having a chirp bandwidth and a start frequency (fSTART), and the at least one scan chirp having a scan bandwidth greater than the chirp bandwidth; transmitting an RF signal representing the frame, wherein the transmit power is zero or reduced to less than the power of a regular chirp during the transmission of the at least one scan chirp;Receiving an RF radar signal (yHF(t)) corresponding to the frame, down-converting the received RF radar signal (yHF(t)) to a baseband using the local oscillator signal (sLO(t)), and generating a digital signal (y[n]) based on the down-converted signal (y(t)); detecting first samples affected by interference in a portion of the digital signal (y[n]) corresponding to the scan chirp, and updating the start frequency (fSTART) for the regular chirps following the scan chirp based on the detected first samples; and detecting one or more radar targets based on the digital signal (y[n]), replacing the portion of the digital signal (y[n]) that does not correspond to the regular chirps with preset samples. Method according to claim 1, wherein the frame contains a regular chirp at the first chirp position. Method according to claim 1 or 2, wherein a scan chirp contains a frequency ramp extending over two or more chirp positions. Method according to any one of claims 1 to 3, wherein a scan chirp includes a frequency ramp followed by a modulation pause. Method according to any one of claims 1 to 4, wherein the part of the digital signal (y[n]) that does not correspond to the regular chirps is replaced by zero padding. A method according to any one of claims 1 to 5, wherein the digital signal (y[n]) contains N sub-sequences corresponding to the N chirp positions of the frame, each sub-sequence containing M samples; and wherein detecting the one or more radar targets comprises: calculating an N×M range Doppler map using M range Fourier transforms and N Doppler Fourier transforms; and detecting the one or more radar targets based on the N×M range Doppler map. A method comprising: generating a frequency-modulated local oscillator signal (sLO(t)) with a plurality of frames of chirps and one or more scan chirps between the frames, which are used for measurement data acquisition; generating an RF signal, wherein the transmit power of the RF signal is reduced or zero during the generation of the at least one scan chirp; receiving an RF radar signal (yHF(t)), down-converting the received RF radar signal (yHF(t)) to a baseband using the local oscillator signal (sLO(t)), and generating a digital signal (y[n]) based on the down-converted signal (y(t)) in a first radar chip (1), wherein the digital signal (y[n]) is composed of multiple sequences, each sequence corresponding to either a chirp of a particular frame or a scan chirp; and repeatedly generating metadata indicating whether the the relevant sequence is affected by disturbances in the first radar chip (1);Transferring the metadata along with the digital signal (y[n]) to another chip (2); repeatedly determining a start frequency (fSTART) for the chirps of the frames based on the metadata in the other chip (2). Method according to claim 7, wherein the metadata for each chirp indicates which part(s) of a frequency band of the chirp in question is / are disturbed by interference. Method according to claim 7 or 8, wherein the metadata allows conclusions to be drawn about sample positions within each sequence of samples that are disturbed by interference. Radar system comprising: a local oscillator configured to generate a frequency-modulated local oscillator signal (sLO(t)) with multiple frames of chirps and one or more scan chirps; at least one transmit channel (TX01, TX02) configured to generate an RF signal (sHF(t)) based on the local oscillator signal (sLO(t)), wherein the transmit power of the RF signal is reduced to zero or below the power of a normal chirp during the generation of the at least one scan chirp;at least one receive channel (RX01, RX02) arranged in a first radar chip (1) configured to receive an RF radar signal (yHF(t)), to downmix the received RF radar signal (yHF(t)) to a baseband using the local oscillator signal (sLO(t)), and to generate a digital signal (y[n]) based on the downmixed signal (y(t)), wherein the digital signal (y[n]) is composed of several sequences, each sequence corresponding either to a chirp of a particular frame or to a scan chirp; a first processor unit arranged in the first radar chip (1) configured to repeatedly generate metadata indicating whether the sequence in question is disturbed by interference; a communication link configured to transmit the metadata together with the digital signal (y[n]) from the first radar chip (1) to a further chip (2);and a second processor unit (55) arranged in the further chip (2), which is configured to repeatedly determine a start frequency (fSTART) for the chirps of the frames based on the metadata.; Radar system according to claim 10, further comprising: a ramp generator (111) coupled to the local oscillator (101) and configured to control the local oscillator (101) according to one or more ramp parameters including the start frequency (fSTART) to cause the local oscillator (101) to generate the frequency-modulated local oscillator signal (sLO(t)) with the multiple frames of chirps and the one or more scan chirps. Radar system according to claim 11, wherein the second processing unit (55) is coupled to the ramp generator (111) and is configured to update the start frequency (fSTART) used by the ramp generator (111). Radar system according to one of claims 10 to 12, wherein the local oscillator (101), the ramp generator (111), the at least one receive channel (RX01, RX02) and the first processing unit are integrated in a monolithic integrated microwave circuit (MMIC) (1), wherein the second processing unit (55) is integrated in a further integrated circuit (2), and wherein the MMIC (1) and the further integrated circuit (2) are connected by a communication link for sending the digital signal (y[n]) and the corresponding metadata from the MMIC (1) to the further integrated circuit (2). Radar system according to claim 13, further comprising: a further connection between the MMIC (1) and the further integrated circuit (2), wherein the second processing unit (55) is configured to update the start frequency (fSTART) used by the ramp generator (111) via the further connection. Radar system according to one of claims 10 to 14, further comprising: a third processing unit (52) configured to detect one or more radar targets based on the digital signal (y[n]), wherein the third processing unit (52) operates independently of the second processing unit (55), wherein the second processing unit (55) is configured to operate independently of and simultaneously with the third processing unit (52).

Images