Interference Suppression in FMCW Radar Systems

Abstract

This document describes techniques configured to identify detections caused by interference signals in the output of a Frequency Modulated Continuous Wave (FMCW) radar system. The detections are detected as being caused by interference sources based on the number of detections assigned to bins in a velocity-direction histogram.

Description

[0001] Related Applications

[0002] This application claims priority to European Patent Application No. EP21179613.1, filed on Jun. 15, 2021. The entire content of this application is incorporated herein by reference. Background Art

[0003] Radar systems are employed in autonomous vehicles (AVs) to identify and track objects within the sensing range of the radar system (e.g., between 1 meter and 250 meters). The radar system is configured to generate detections within corresponding time detection windows, where each detection corresponds to a point (within the range of the radar system) in the environment of the radar system at which the radar system determines that a target exists. Thus, for each detection, the radar system identifies the position of the detection relative to the radar system. Additionally, the radar system is configured to calculate a corresponding velocity value (e.g., the velocity of the target relative to the radar system) for each detection and is further configured to calculate a corresponding direction value (e.g., the direction of the target relative to the radar system) for each detection.

[0004] An exemplary type of radar system is a frequency-modulated continuous-wave (FMCW) radar system, where the FMCW radar system is configured to transmit an FMCW signal including an FMCW chirp into the environment. The FMCW radar system is further configured to detect radar signals having frequencies within a predefined spectrum; when the detected radar signal includes the reflection of the FMCW chirp from a target, the radar system outputs a detection over time based on the difference between the frequency of the local oscillator (LO) and the frequency of the detected radar signal, where the distance of the radar system to the target is based on the difference between the frequency of the LO and the frequency of the detected radar signal. More specifically, the radar system generates detections based on the detected radar signal downmixed with the local oscillator (LO), where in one example the LO corresponds to the transmitted FMCW signal. The radar system can calculate the velocity of the detection based on the phase of the downmixed signal and can employ beamforming techniques to calculate the direction of the detection. For the corresponding time detection window, the radar system can generate a multitude of detections (where the number of detections can depend on the range of the radar system, the number of transmit and receive antennas, and other factors).

[0005] As more and more AVs cross the road, and as radar systems become more prevalent, it is envisioned that an FMCW radar system can generate false detections (not detections caused by the reflection of an FMCW chirp from a target) due to the radar signal being transmitted into the environment by a radar radiation source. In other words, when there is actually no target at a certain position in the environment, the radar system may output data indicating the presence of a target at that position.

[0006] There are several conventional ways to identify a detected radar signal as being emitted by an interference source and / or avoid detection of an interference signal. For example, when the radar system is an FMCW radar system, the radar system can use interference nulling to detect FMCW chirps emitted by an interference source (e.g., another radar system). However, interference nulling has several drawbacks, including the elimination of potentially useful information from the output of the analog-to-digital converter (ADC). Additionally, when using interference nulling, it can be difficult to detect interference signals in the original data stream. Another conventional way is to randomly change the modulation parameters of the FMCW radar system (e.g., modify the phase of the FMCW chirp). However, this introduces additional circuitry and processing complexity. To avoid detection of interference signals, the FMCW radar system can use frequency hopping when emitting the FMCW signal, such that the radar system can emit a first FMCW chirp within a first baseband spectrum and later emit a second FMCW chirp within a second baseband spectrum (to avoid interference signals having frequencies within the first baseband spectrum). However, frequency hopping also introduces additional processing complexity and requires the use of additional bandwidth. SUMMARY OF THE INVENTION

[0007] The following is a brief overview of the subject matter described in greater detail herein. This overview is not intended to limit the scope of the claims.

[0008] This disclosure describes various techniques for the detection of markers generated by a frequency-modulated continuous-wave (FMCW) radar system that are due to interference signals (radar signals emitted from a radiation source) rather than reflections of FMCW signals emitted by the radar system. The radar system is configured to generate an FMCW signal that includes FMCW chirps and emit such FMCW signals into the environment of the radar system. The radar system additionally generates a local oscillator (LO) corresponding to the emitted FMCW signal. Within a time detection window, the radar system generates a number of detections (e.g., in 2-dimensional space or 3-dimensional space), where each detection is generated based on a down-converted signal produced by mixing the LO with a radar signal detected by the radar system. Thus, the number of detections represents points in the scene at which the radar system has determined that one or more targets are present.

[0009] The radar system is configured to calculate corresponding velocity values and direction values for the detections. The velocity value for a detection represents the calculated velocity of a target that the radar system perceives to be in the scene. The direction value for the detection represents the calculated direction of the target that the radar system perceives to be in the scene relative to the radar system. As mentioned above, however, the detections can be due to an interference source (emitting a continuous-wave (CW) radar signal) rather than reflections of the FMCW signal, and thus there may not be any targets in the scene corresponding to the detections.

[0010] The radar system is further configured to perform velocity gating and direction gating for the detections based on the velocity value and the direction value calculated for the detections. More specifically, the radar system may be able to calculate a velocity in a range between a first value and a second value (e.g., between 0 m / s and 64 m / s). The radar system may perform a gating operation to place the calculated velocity within one of a plurality of discrete non-overlapping gates, where a gate represents a sub-range within the range. Thus, when the velocity for a detection is 54.2343 m / s, the radar system may assign the detection to a velocity gate that covers the sub-range 54 m / s – 55 m / s. For the direction calculated for a detection, the radar system begins a similar process (e.g., assigning the detection to a direction gate).

[0011] The radar system constructs a velocity-direction (multi-dimensional) histogram for each time detection window based on the velocity gates and the direction gates, where the histogram includes a number of non-overlapping bins, where each bin corresponds to a respective velocity gate and a respective direction gate. For example, when there are 64 velocity gates and 64 direction gates, the velocity-direction histogram includes 4096 bins (e.g., [VG1,BG1], [VG2,BG1], [VG3,BG1], ……, [VG63,BG64], [VG64,BG64]). Detections assigned to the same velocity gate and the same direction gate are assigned to the same bin in the velocity-direction histogram.

[0012] The radar system uses the histogram to identify detections caused by interference signals during the time detection window. More specifically, the radar system is configured to identify detections caused by interference signals during the time detection window based on the number of detections assigned to the bins in the histogram. Even more specifically, and as will be described in more detail herein, the radar system is configured to determine the number of detections assigned to each bin and is further configured to determine the number of detections assigned to bins that are a neighborhood of the bin in the histogram. The radar system identifies detections caused by interference signals based on the number of detections assigned to the bin and the aggregated number of detections assigned to the neighborhood bins. This is possible due to the radar system's downmixing of the local oscillator (LO) (including the FMCW chirp) and the detected signal; when the detected signal is a continuous wave (CW) signal, the downmixed signal includes a chirp in the baseband spectrum of the radar system, where the chirp has a similar phase shift. When a fast Fourier transform is performed on these chirps, energy exists over a wide range of distances along a single direction. Additionally, the similar phase shift indicates a constant velocity – thus, when there are a large number of detections with the same velocity and direction, these detections are likely caused by interference signals.

[0013] In some embodiments, the radar system is configured to employ a number of thresholds to identify detections caused by interference signals (as opposed to reflections of the FMCW signal). In one example, the radar system is configured to determine the number of detections in a bin of the histogram and is further configured to compare the number of detections to a first threshold to determine if the detections in the bin are likely caused by an interference signal. When the number of detections is greater than the first threshold, the radar system is configured to compare the number of detections to a second (higher) threshold. When the number of detections in the bin is greater than the second threshold, the radar system marks each detection assigned to the bin as being caused by an interference signal such that the detections can be filtered from the output data. When the number of detections is less than the second threshold (but greater than the first threshold), the radar system is configured to compare the number of detections assigned to the bin to the aggregated number of detections assigned to bins that are a neighborhood of the bin (e.g., where the bins that are a neighborhood can be identified a priori).

[0014] When the difference between the number of detections assigned to the bin and the aggregated number of detections assigned to the neighborhood bins is above a threshold, each detection assigned to the bin can be marked as being caused by an interference signal. Additionally, when the aggregated number of detections assigned to bins that are an extended neighborhood of the bin is below a threshold (e.g., there is a sparse number of detections with similar speed and direction), the detections assigned to the neighborhood bins can each be marked as being caused by the interference signal. The radar system can filter such detections from the output data after marking the detections as being caused by the interference signal such that the output data generated (and transmitted to, e.g., a computing system of an autonomous vehicle for further processing) by the radar system does not include the detections marked as being caused by the interference signal. In one example, an autonomous vehicle performs driving maneuvers based on the output data generated by the radar system (where the detections are marked as being caused by the interference signal from which they are filtered).

[0015] A brief summary is presented above in order to provide a basic understanding of some aspects of the systems and / or methods described herein. This summary is not an extensive overview of the systems and / or methods described herein. This summary is not intended to identify key / critical elements or to delineate the scope of such systems and / or methods. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented below. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a schematic illustration of an autonomous vehicle having a frequency modulated continuous wave (FMCW) radar system of an autonomous vehicle, where the radar system is configured to identify detections caused by interference signals.

[0017] Figure 2Is a graph of an exemplary detection generated by an FMCW radar system.

[0018] Figure 3 Illustrates two graphs of the local oscillator (LO) versus time with respect to a reflected signal and a continuous wave (CW) interference signal.

[0019] Figure 4 Is a schematic diagram of a radar system configured to identify detections caused by interference signals.

[0020] Figure 5 Is an exemplary velocity - direction histogram.

[0021] Figure 6 Is a flowchart of a method for identifying that a detection generated by an FMCW radar system is caused by an interference signal.

[0022] Figure 7 Is a flowchart of a method for labeling a detection generated by an FMCW radar system as being caused by an interference signal.

[0023] Figure 8 Is a computing system that may be included in a radar system and / or an autonomous vehicle (AV). Detailed Description

[0024] Now referring to the drawings, various techniques are described regarding a frequency - modulated continuous - wave (FMCW) radar system configured to label detections as being caused by interference signals, such as continuous - wave (CW) signals, where like reference numerals are used throughout to refer to like elements. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more aspects. However, such aspects may clearly be practiced without these specific details. In other instances, well - known structures and devices are shown in block diagram form in order to facilitate describing one or more aspects. Additionally, it should be understood that functionality described as being performed by certain system modules may be performed by multiple modules. Similarly, for example, a module may be configured to perform functionality described as being performed by multiple modules.

[0025] Furthermore, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or". That is, unless otherwise indicated or clear from the context, the phrase "X employs A or B" is intended to mean any natural inclusive permutation. That is, the phrase "X employs A or B" is satisfied by any of the following cases: X employs A; X employs B; or X employs both A and B. Additionally, unless otherwise indicated or clear from the context to be singular, the articles "a" and "an" in this application and the appended claims generally should be understood to mean "one or more".

[0026] This document describes an FMCW radar system configured to identify detections caused by interference signals (as opposed to FMCW signals transmitted by the radar system and reflected from targets) generated by the radar system. The radar system is configured to identify detections caused by interference signals by using a velocity-direction histogram. As will be described in more detail herein, the radar system assigns each detection to a bin among a plurality of bins in the velocity-direction histogram, where the radar system assigns the detection to the bin based on the velocity assigned to the detection and one or more directions assigned to the detection. The radar system is configured to count the number of detections assigned to the bins in the histogram and is further configured to label the detections as being caused by interference signals based on the number of detections assigned to the bins in the histogram. The radar system described herein is an improvement over conventional radar systems in that (unlike radar systems that employ interference nulling) the radar system can identify detections caused by interference signals. Additionally, the radar system does not need to randomly change modulation parameters to identify detections caused by interference signals; furthermore, the radar system does not need to identify the baseband frequency of the FMCW signals transmitted by the radar system to identify detections caused by interference signals.

[0027] Now referring to Figure 1 , a schematic diagram depicting an autonomous vehicle 100 navigating in an environment is illustrated. In Figure 1 the example depicted, the environment surrounding the autonomous vehicle (AV) 100 includes a target 102 (such as a telephone pole, a pedestrian, a car, a building, or any other suitable object that may be in the environment with the AV 100). Additionally, the environment includes an interference source 104, where the interference source 104 emits an interfering radar signal in the environment. For example, the interference source 104 is a radar system coupled to another device (e.g., another AV). Furthermore, the interference signal can be a continuous wave (CW) radar signal having a frequency between 76 GHz and 81 GHz.

[0028] The AV 100 includes a radar system 106, where the AV 100 performs driving maneuvers based on the output of the radar system 106. The radar system 106 is an FMCW radar system that emits an FMCW signal including a plurality of chirps into the environment. The radar system 106 detects a return signal that has frequencies in its spectrum that cover the frequencies in the chirps emitted by the radar system 106. The chirps include up-chirps and / or down-chirps, where during an up-chirp the frequency of the FMCW signal increases (e.g., linearly) over time and during a down-chirp the frequency of the FMCW signal decreases (e.g., linearly) over time. The radar system 106 emits an FMCW radar signal 108 including chirps into the environment and towards a target 102. The return signal 110 is detected by the radar system 106, where the return signal 110 is a reflection of the radar signal 108 from the target 102. The radar system 106 generates a detection based on the return signal 110.

[0029] As Figure 1 As a further illustration, an interference source 104 emits an interference signal 112, where the interference signal 112 travels towards the radar system 106. The radar system 106 detects the interference signal 112 and generates a detection based on the interference signal.

[0030] More specifically, with respect to the radar system 106 that generates detections, the radar system 106 generates a local oscillator (LO); in one example, the LO corresponds to the emitted FMCW signal 108. The radar system 106 detects the radar signal and down-converts the detected radar signal with the LO, thereby forming a down-converted signal. When the detected signal is a reflection of the emitted FMCW signal from the target 102, the frequency of the down-converted signal represents the distance between the radar system 106 and the target 102, and the radar system 106 generates a detection corresponding to the target. However, when the detected signal is an interference signal, the down-converted signal includes a frequency chirp instead of a relatively constant frequency; when a fast Fourier transform is performed on the down-converted signal, the energy spreads over a plurality of distances, and thus the radar system 106 generates a number of detections. However, the detections do not correspond to any target in the environment.

[0031] The radar system 106 includes processing circuitry 114 that is configured to generate detections based on the LO and the detected radar signals and is further configured to assign values to the detections, such as range values, velocity values, and direction values. As will be described below, the processing circuitry 114 is also configured to perform velocity gating and direction gating on the detections. The processing circuitry 114 includes an interference source tagger module 116 that is configured to identify detections caused by interference signals (e.g., interference signal 112 emitted by interference source 104) generated by the processing circuitry 114 and is configured to label such detections as being caused by the interference signal. With respect to gating and as to velocity values, the processing circuitry 114 may be configured to accurately calculate the velocity for detections within a specific range (e.g., 0 m / s to 50 m / s). The processing circuitry 114 may divide the range into non-overlapping gates, where each gate corresponds to a sub-range of the range and where the gates collectively cover the range. For example, the range may be divided into 50 sub-ranges, where each sub-range covers a 1 m / s sub-range. However, it should be understood that the gates need not cover equivalent sub-ranges, and more or fewer than 50 gates are contemplated. The processing circuitry 114 performs a similar process when performing direction gating (where the direction gating may be in two or three dimensions). Thus, the detections are assigned velocity gates and one or more direction gates based on the velocity value calculated for the detections and the direction value calculated for the detections.

[0032] The interference source tagger module 116 is configured to construct a (two-dimensional) velocity-direction histogram based on the velocity gates and direction gates cited above. For example, when there are 64 velocity gates and 64 direction gates, the histogram includes 4096 bins (e.g., (e.g., [VG1,BG1], [VG2,BG1], [VG3,BG1], ……, [VG63,BG64], [VG64,BG64])). The interference source tagger module 116 assigns the detections to the bins in the histogram based on the velocity gates and direction gates assigned to the detections. Thus, detections having the same velocity gates and direction gates assigned to them are in turn assigned to the same bin in the histogram. Conversely, two detections having different velocity gates and / or direction gates assigned to them are assigned by the interference source tagger module 116 to two different bins in the histogram.

[0033] The interference source labeler module 116 labels detections as being caused by interference signals based on the number of detections assigned to bins in a histogram. As will be described in more detail below, the interference source labeler module 116 may utilize various thresholds to determine whether a detection assigned to a bin is caused by an interference signal. In other embodiments, the interference source labeler module 116 may employ machine learning techniques in combination with identifying detections caused by interference signals. For example, a neural network (such as a deep neural network (DNN), a recurrent neural network (RNN), etc.) may be provided with a histogram and may identify detections caused by interference signals based on the number of detections assigned to bins in the histogram. More specifically, the machine learning techniques may be trained by labeled training data, where the labeled training data includes a histogram and a label indicating whether the detection assigned to the detection in the histogram is caused by an interference signal.

[0034] Now referring to Figure 2 , a graph 200 is presented that depicts detections generated by the radar system 106 (before filtering detections caused by the interference signal 112 emitted by the interference source 104) during a time detection window. As previously noted, these detections represent points in space relative to the radar system 106 where the radar system 106 believes that the FMCW signal emitted by the radar system 106 has reflected from a target and returned to the radar system 106. However, in fact, the detections illustrated in region 202 of graph 200 are caused by interference signals and thus do not represent one or more targets in the environment. As noted above, the radar system 106 assigns a velocity value to each of the detections illustrated in graph 200. When the interference signal 112 emitted by the interference source 104 is, for example, a CW signal, a set of detections having a similar velocity and located in a direction relative to the radar system 106 are generated by the radar system 106 (e.g., the detections caused by the interference signal in region 202 have a similar velocity). The detections having the same / similar direction and the same / similar velocity are the effect of the analog mixing of the local oscillator signal with, for example, a CW interference signal.

[0035] Figure 3 Depicts a pair of graphs 300 and 302. The first graph 300 illustrates a pair of chirps (chirp n and chirp n + 1) emitted by the radar system 106 into the environment (where the emitted chirps are also LOs). The first graph 300 also illustrates the radar signals detected by the radar system 106, where the detected radar signals are reflections of the emitted chirps. It can be determined that the reflection of the chirp detected by the radar system 106 has a frequency included in the baseband spectrum and is timely offset from the chirp. Therefore, when the LO is mixed down with the received radar signal, a constant value is obtained for most of the length of the chirp (where the constant value is the difference between the frequency of the chirp emitted over time and the frequency of the detected chirp).

[0036] Conversely, the second graph 302 illustrates the transmitted FMCW signal emitted by the radar system 106, where the FMCW signal includes a pair of chirps. The second graph 302 also illustrates the interfering CW signal emitted by the interfering source 104. Downmixing of the LO with the interfering CW signal results in a downmixed signal that includes chirps in the baseband spectrum. A fast Fourier transform (FFT) performed on the downmixed signal causes the radar system 106 to generate numerous detections with a wide range of distances. Additionally, in the chirps of the downmixed signal, the phase shift is the same as the chirp changes, indicating a constant velocity across the generated detections. As will be described in more detail below, these features allow for a difference between detections associated with the reflected return signal and detections caused by the interfering signal.

[0037] Now refer to Figure 4 , a functional block diagram of the radar system 106 is illustrated. The radar system 106 includes a signal generator 402 that generates an FMCW signal including a plurality of chirps. A power splitter 404 receives the FMCW signal and splits the FMCW signal (e.g., duplicates the FMCW signal). The radar system 106 also includes a transmit antenna 406 that receives a first instance of the FMCW signal from the power splitter 404 and emits the first instance of the FMCW signal into the environment of the radar system 106. The radar system 106 also includes a downmixer 408, where the downmixer 408 receives a second instance of the FMCW signal from the power splitter 404. The first instance of the FMCW signal has a higher power than the second instance of the FMCW signal because the power of the first instance of the FMCW signal must be sufficient to allow the transmit antenna 406 to transmit the FMCW signal into the environment with sufficient power to allow detections within the desired range.

[0038] The radar system 106 also includes a receive antenna 410 that detects radar signals from the environment of the radar system 106. The detected radar signal can be or include a reflection of the FMCW signal emitted by the transmit antenna 406 (when the FMCW signal is reflected from a target in the environment of the radar system 106). Additionally, the detected radar signal can be or include an interfering signal emitted from the interfering source 104. Although the radar system 106 is illustrated as including a transmit antenna 406 and a receive antenna 410, it should be understood that the radar system 106 includes a plurality of transmit antennas and a plurality of receive antennas such that beamforming can be initiated.

[0039] Amplifier 412 is operatively coupled to receive antenna 410 and amplifies the detected signal output by receive antenna 410, thus outputting an amplified signal. Downconverter 408 is electrically coupled to amplifier 412 and downconverts the LO output by power splitter 404 and the amplified signal output by amplifier 412. Downconverter 408 outputs a downconverted analog signal that identifies the frequency difference over time between the LO and the amplified signal output by amplifier 412.

[0040] Radar system 406 further includes low pass filter 414, which is electrically coupled to downconverter 408. Low pass filter 414 is configured to filter out higher frequencies from the downconverted signal and output a filtered signal, where the filtered signal includes frequencies included in the baseband spectrum. Second amplifier 416 is optionally electrically coupled to low pass filter 414, where second amplifier 416 amplifies the filtered signal output by low pass filter 414. Radar system 406 further includes analog-to-digital converter (ADC) 418, which receives the amplified signal output by second amplifier 416 and converts such signal into a digital signal.

[0041] Radar system 106 further includes processing circuit 114, where processing circuit 114 is configured to receive the digital signal output by ADC 418. In one example, processing circuit 114 is a digital signal processor (DSP). In another example, processing circuit 114 is or includes a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a graphics processing unit (GPU), a microcontroller, or other suitable processing circuit.

[0042] Processing circuit 114 includes distance calculation module 420, speed calculation module 422, and direction calculation module 424. Processing circuit 114 generates a detection of the environment based on the digital signal output by ADC 418. Distance calculation module 420 is configured to calculate the distance relative to radar system 106 for the detection and is optionally further configured to perform range gating. Speed calculation module 422 is configured to calculate the speed for the detection and is further configured to perform speed gating for the detection. Direction calculation module 424 is configured to calculate the direction relative to radar system 106 for the detection and is further configured to perform direction gating for the detection.

[0043] More specifically, the radar system 106 is configured to unambiguously detect certain speed ranges; such ranges can be divided into a number (e.g., 64) of discrete non-overlapping bins, and the speed calculation module 422 can assign each detection to a speed bin based on the speed assigned to the detection by the speed calculation module 422. Similarly, the radar system 106 is configured to unambiguously detect certain direction ranges. This direction range can be divided into, for example, 64 discrete non-overlapping bins, and each detection is assigned to an appropriate direction bin based on the direction calculated by the direction calculation module 424 for the detection.

[0044] The processing circuit 114 further includes an interference source tagger module 116, which is configured to identify and tag detections caused by interference signals (e.g., emitted by the interference source 104) such that the tagged detections can be filtered. To this end, the interference source tagger module 116 includes a histogram generator module 426, which constructs a speed-direction histogram based on the speed and direction bins as described above. The histogram generator module 426 assigns detections to bins in the histogram based on the speed bin and direction bin to which the detections have been assigned. Thus, detections assigned to the same bin in the histogram are each assigned to the same speed bin and the same direction bin respectively by the speed calculation module 422 and the direction calculation module 424.

[0045] Briefly turning to Figure 5 , an exemplary speed-direction histogram 500 is illustrated. The height of the bins in the histogram 500 represents the number of detections assigned to the bins in the histogram 500 within the time detection window. Thus, as illustrated by the height of bin 502, there is a relatively large number of detections assigned to bin 502 in the histogram 500.

[0046] Returning to Figure 4 , the interference source tagger module 116 further includes a filtering module 428, which is configured to mark detections as being caused by interference signals and is also configured to filter such detections from the output signal generated by the radar signal 106. Generally, the filtering module 428 is configured to count the number of detections in the bins of the histogram and mark detections as being caused by interference signals based on the number of detections in the bins.

[0047] Referring to Figure 6 and Figure 7 , a method performed by the radar system 106 is illustrated. Although the method is shown and described as a series of actions performed in sequence, it should be understood and appreciated that the method is not limited by the order of the sequence. For example, some actions may occur in an order different from that described herein. Additionally, an action may occur concurrently with another action. Further, in some cases, not all actions are required to implement the method described herein.

[0048] In addition, some of the actions described herein may be computer-executable instructions that can be implemented by one or more processors and / or stored on one or more computer-readable media. The computer-executable instructions can include routines, subroutines, programs, execution threads, and the like. Further still, the results of the actions of the method can be stored in a computer-readable medium, displayed on a display device, and so on.

[0049] Referring separately to Figure 6 , method 600 is illustrated, which is performed by radar system 106 in combination with detecting markers as being caused by interference signals and filtering such detections from the output data. Method 600 begins at 602, and at 604, an LO signal corresponding to the transmitted FMCW radar signal (where the FMCW radar signal includes a number of chirps) is generated. At 606, the LO signal is mixed down with the received radar signal to generate a mixed-down signal. At 608, a velocity-direction histogram is generated based on the mixed-down signal, where each detection in the time detection window is assigned to a bin in the histogram based on the velocity calculated for the detection and the direction calculated for the detection. As previously indicated, velocity gating and direction gating can be performed in combination with binning the detections in the velocity-direction histogram.

[0050] At 610, detections are identified as being caused by interfering radar signals (e.g., interfering CW radar signals) based on the histogram. At 612, output data identifying detections of the scene is generated, where detections marked as being caused by interference signals are not included or otherwise represented in the output data. Method 600 completes at 614.

[0051] Now referring to Figure 7 , a flowchart is illustrated that illustrates method 700 performed by filtering module 428 in combination with marking detections as being caused by interference signals. Method 700 begins at 702, and at 704, bins are selected from the histogram and the number of detections (N) in the bins of the histogram is counted. At 706, a determination is made as to whether N is greater than a first predefined threshold (threshold 1). If N is not greater than the first predefined threshold, method 700 proceeds to 708, where a determination is made as to whether there are more bins to process. When it is determined at 708 that there are more bins to process, method 700 returns to 704, where another bin in the histogram is selected and the number of detections in the selected bin is counted.

[0052] When it is determined at 706 that N is greater than a first predefined threshold, method 700 proceeds to 710, where the number of detections to select (NB1) in a first subset of neighborhood bins is determined. The bins that are neighborhood bins of the selected bins can be defined based on the operating mode of radar system 106. In one example, a neighborhood bin is a bin immediately adjacent to the selected bin in a histogram. In another example, the first subset of neighborhood bins can include the neighborhood bin having the second highest count among all neighborhood bins and the neighborhood bin having the fourth highest count among all neighborhood bins. The bins to be included in the first subset of neighborhood bins can be determined empirically.

[0053] At 712, a determination is made as to whether N is greater than a second predefined threshold (threshold 2), where the second threshold is higher than the first threshold. When N is greater than the second predefined threshold, method 700 proceeds to 714, where each detection in the bin is marked as being caused by an interference signal. When N is not greater than the second threshold, method 700 proceeds to 716, where a determination is made as to whether the difference between N and NB1 exceeds a third predefined threshold. In one example, the third threshold can be equivalent to the first threshold. When it is determined at 716 that the difference between N and NB1 exceeds the third threshold, method 700 proceeds to 714, where each detection in the bin is marked as being caused by an interference signal. When it is determined at 716 that the difference between N and NB1 does not exceed the threshold, the method proceeds to 708, where a determination is made as to whether there are more bins to select.

[0054] When each detection in the bin is marked as being caused by an interference signal at 714, method 700 proceeds to 718, where the number of detections (NB2) in a second subset of neighborhood bins is determined. Just as with the first subset of neighborhood bins, the bins to be included in the second subset of neighborhood bins can be determined empirically. At 720, a sparsity check is made. More specifically, a determination is made as to whether there is a sparse number of detections around the bins that have been identified as corresponding to interference signals. At 720, a determination is made as to whether NB2 is less than a fourth predefined threshold. When NB2 is not less than the fourth threshold, method 700 proceeds to 708. When it is determined at 720 that NB2 is less than the fourth threshold, then method 700 proceeds to 722, where each detection in each of the neighborhood bins is identified as being caused by an interference signal. This can occur due to defects in velocity gating and / or direction gating because velocities near the gate boundaries may fall into different gates.

[0055] When each detection in each neighborhood bin is marked as being caused by an interference signal, method 700 returns to 708, where a determination is made as to whether there are more bins to process. When there are no more bins to process, method 700 ends at 724.

[0056] Regarding method 700, the threshold can be determined empirically and may vary depending on the radar system employed, the mode of the radar system, etc. Additionally, the bins in the histogram are considered to be periodic.

[0057] Although method 700 has been described as using a threshold to identify detections caused by interference signals, it should be understood that other approaches are envisioned. For example, as cited above, machine learning techniques can be employed, where the machine learning technique is provided with a velocity - direction histogram as an input, and the machine learning technique outputs an indication as to which detections are caused by interference signals based on the content of the histogram.

[0058] Now referring to Figure 8 , an advanced illustration of an exemplary computing device 800 that can be used in accordance with the systems and methods disclosed herein is shown. For example, computing device 800 can be used in a system configured to control the mechanical systems of an autonomous vehicle. By another example, computing device 800 can be used in a system configured to identify detections caused by interference signals. Computing device 800 includes at least one processor 802 that executes instructions stored in memory 804. The upright can be, for example, instructions for implementing the functionality described as being performed by one or more of the components discussed above or for implementing one or more of the amplifications described above. Processor 802 can access memory 804 via system bus 806. In addition to storing executable instructions, memory 804 can also store cross - correlated outputs, ultrasonic sensor outputs, calibration settings, parameters, etc.

[0059] Computing device 800 additionally includes a data storage area 808 that can be accessed by processor 802 via system bus 806. Data storage area 808 can include executable instructions, thresholds, histograms, etc. Computing device 800 also includes an input interface 810 that allows external devices to communicate with computing device 800. For example, input interface 810 can be used to receive instructions from an external computer device, from a user, etc. Computing device 800 also includes an output interface 812 that interfaces computing device 800 with one or more external devices. For example, computing device 800 can display text, images, etc. via output interface 812.

[0060] Additionally, although illustrated as a single system, it should be understood that computing device 800 can be a distributed system. Thus, for example, several devices can communicate via a network connection and can jointly perform the tasks described as being performed by computing device 800.

[0061] The various functions described herein can be implemented in hardware, software, or any combination thereof. If implemented in software, the functions can be stored on or transmitted via a computer-readable medium as one or more instructions or code. A computer-readable medium includes a computer-readable storage medium. A computer-readable storage medium can be any available storage medium accessible by a computer. By way of example, and not limitation, such computer-readable storage medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and that is accessible by a computer. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, and Blu-ray disc (BD), where disks typically reproduce data magnetically, and discs typically reproduce data optically with lasers. In addition, a propagated signal is not included within the scope of computer-readable storage media. A computer-readable medium also includes a communication medium that includes any medium that facilitates transfer of a computer program from one place to another. A connection, for example, can be a communication medium. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of communication medium. Combinations of the above should also be included within the scope of computer-readable media.

[0062] Alternatively or in addition, the functionality described herein can be performed, at least in part, by one or more hardware logic components. By way of example, and not limitation, illustrative types of hardware logic components that can be used include field programmable gate arrays (FPGA), application specific integrated circuits (ASIC), application specific standard products (ASSP), systems-on-a-chip (SOC), complex programmable logic devices (CPLD), etc.

[0063] The features described herein relate to detection caused at least in part by interference signals, at least according to the examples provided below.

[0064] (A1)In one aspect, some embodiments include a method performed by processing circuitry of a radar system of an AV, where the method includes: calculating, for a time detection window and based on a frequency-modulated continuous wave (FMCW) signal emitted by the radar system: 1) the speed of a detection in a scene; and 2) the direction of the detection relative to the radar system. The method further includes: generating a speed-direction histogram of the detection window, where the speed-direction histogram includes bins, and further where the detection is assigned to a bin in the bins based on the speed assigned to the detection and the direction of the detection relative to the radar system. The method further includes: identifying the detection as being caused by an interference signal based on the number of detections in the bin of the speed-direction histogram assigned to the detection window. The method additionally includes: generating output data based on the FMCW signal, the output data identifying the detections in the scene, where the detections identified in the output data do not include the detections identified as being caused by the interference signal, and where the AV performs driving maneuvers based on the output data.

[0065] (A2)In some embodiments of the method as described in A1, the speed and direction are calculated for a number of detections in the time detection window based on the FMCW signal, where the number of detections are assigned to corresponding bins in the speed-direction histogram of the time detection window.

[0066] (A3)In some embodiments of the method as described in A2, each detection assigned to the bin among the number of detections is identified as being caused by the interference signal based on the number of detections in the bin of the speed-direction histogram assigned to the detection window.

[0067] (A4)In some embodiments of the method as described in any one of A1 to A3, identifying the detection as being caused by the interference signal includes calculating the number of detections in the bin of the speed-direction histogram. Identifying the detection as being caused by the interference signal further includes determining that the number of the detections exceeds a threshold, and where the detection is identified as being caused by the interference signal based on the number of the detections exceeding the threshold.

[0068] (A5) In some embodiments of the method as described in A4, identifying the detection as being caused by the interference signal further includes: a) calculating the number of the detections assigned to the bin in the velocity - direction histogram; b) calculating the corresponding number of detections for the bins adjacent to the bin in the velocity - direction histogram; c) selecting a subset of bins from the bins adjacent to the bin; d) determining the number of detections in the subset of bins; and e) calculating the difference between the number of the detections assigned to the bin in the velocity - direction histogram and the number of the detections in the subset of bins, wherein the detection is identified as being caused by the interference signal based on the difference between the number of the detections assigned to the bin and the number of the detections in the subset of bins.

[0069] (A6) In some embodiments of the method as described in A5, identifying the detection as being caused by the interference signal further includes: f) determining that the difference between the number of the detections assigned to the bin in the velocity - direction histogram and the number of the detections in the subset of bins exceeds a second threshold, wherein the detection is identified as being caused by the interference signal based on the difference between the number of the detections assigned to the bin and the number of the detections in the subset of bins exceeding the second threshold.

[0070] (A7) In some embodiments of the method as described in any one of A5 to A6, each detection assigned to the bin is identified as being caused by the interference signal, and wherein the output data does not include each detection assigned to the bin.

[0071] (A8) In some embodiments of the method as described in any one of A5 to A7, the method further includes: identifying that a second detection is caused by the interference signal, wherein the detection is assigned to a second bin adjacent to the bin, and wherein identifying that the second detection is caused by the interference signal includes: a) selecting a second subset of bins from the bins adjacent to the bin; b) determining the number of detections in the second subset of bins; c) comparing the number of the detections in the second subset of bins with the second threshold; and d) determining that the number of the detections in the second subset of bins is below the second threshold, wherein the second detection is identified as being caused by the interference signal based on the number of the detections in the second subset of bins being below the second threshold.

[0072] (A9) In some embodiments of the method as described in A8, each detection adjacent to the bin in each bin is marked as being caused by the interference signal based on the number of the detections in the second subset of bins being below the second threshold.

[0073] In some embodiments of the method according to any one of A1 to A9, the computing system communicates with the radar system, wherein the computing system is configured to identify an object in the scene based on the output data.

[0074] (B1) In another aspect, some embodiments include a method performed by a radar system of an AV. The method includes: receiving a signal in a scene; and generating a detection based on the received signal and a frequency-modulated continuous wave (FMCW) local oscillator within a time detection window, wherein the detection is assigned a speed value and a direction value. The method further includes: assigning the detection to a bin in a speed-direction histogram for the time detection window based on the speed value and the direction value assigned to the detection, wherein the speed-direction histogram includes a plurality of bins. The method additionally includes: determining the number of detections in the bin; and identifying that the signal is emitted by an interference source based on the number of detections in the bin. The method additionally includes: after identifying that the signal is emitted by the interference source, generating output data including detections in the time detection window, wherein the detections do not include the detections since the signal is identified as being emitted by the interference source, and further wherein the AV performs a driving maneuver based on the output data.

[0075] (B2) In some embodiments of the method according to B1, the method further includes: determining that the number of detections in the bin is greater than a predefined threshold, wherein the signal is identified as being emitted by the interference source based on the number of detections in the bin being greater than the predefined threshold.

[0076] (B3) In some embodiments of the method according to any one of B1 to B2, the bin includes a plurality of detections assigned to the bin, wherein the plurality of detections are assigned the speed value and the direction value, wherein the plurality of detections are identified as being caused by the signal emitted by the interference source, and further wherein the output data does not include the plurality of detections.

[0077] (B4) In some embodiments of the method according to any one of B1 to B3, the method further includes: determining the number of detections in a second bin among the plurality of bins for the time window, wherein the signal is identified as being emitted by the interference source based on the number of detections in the second bin.

[0078] In some embodiments of the method according to any one of B1 to B4, the method further includes: a) defining bins in the velocity - direction histogram as neighboring bins of the bins in the velocity - direction histogram; b) determining the number of detections in a subset of the bins in the neighboring bins; and c) calculating the difference between the number of detections in the subset of the bins and the number of detections in the bin, wherein the signal is identified as being emitted by the interference source based on the difference between the number of detections in the subset of the bins and the number of detections in the bin.

[0079] In some embodiments of the method according to B5, the method further includes: determining that the difference between the number of detections in the subset of the bins and the number of detections in the bin is greater than a threshold, wherein the signal is identified as being emitted by the interference source based on the difference between the number of detections in the subset of the bins and the number of detections in the bin being greater than the threshold.

[0080] In another aspect, some embodiments include a method performed by a radar system including an antenna configured to detect radar signals and a processing circuit. The method includes: calculating, for a time detection window and based on the radar signals: the velocity of detections in a scene; and the direction of the detections relative to the radar system. The method further includes: generating a velocity - direction histogram of the detection window, wherein the velocity - direction histogram includes bins, and further wherein the detections are assigned to bins in the histogram based on the velocity assigned to the detections and the direction of the detections relative to the radar system. The method also includes: identifying the detections as being caused by an interference source based on the number of detections in the bins of the velocity - direction histogram assigned to the detection window. The method additionally includes: generating output data based on the radar signals, the output data identifying the detections in the scene, wherein the detections identified in the output data do not include the detections identified as being caused by the interference source.

[0081] In some embodiments of the method according to C1, an AV is attached to the radar system, wherein the AV performs driving maneuvers based on the output data.

[0082] In some embodiments of the method according to any one of C1 to C2, a transmitter is configured to transmit a frequency - modulated continuous - wave (FMCW) signal, wherein the output data is further based on the FMCW signal.

[0083] (C4)In some embodiments of the method as described in C3, speed and direction are calculated for a number of detections in the time detection window based on the FMCW signal, where the number of detections are assigned to a number of bins in the speed - direction histogram of the detection window, and further where each detection assigned to a bin among the number of detections is identified as being caused by the interference source based on the number of detections assigned to the bin in the speed - direction histogram of the detection window.

[0084] (D1)In another aspect, some embodiments include an AV including a radar system having a processing circuit, where the processing circuit is configured to perform one or more of the above - described methods (e.g., any one of A1 to A10, B1 to B6, and C1 to C4).

[0085] (E1)In another aspect, some embodiments include a radar system including a processing circuit, where the processing circuit is configured to perform one or more of the above - described methods (e.g., any one of A1 to A10, B1 to B6, and C1 to C4).

[0086] (F1)In yet another aspect, some embodiments of a radar system include an antenna configured to detect radar signals. The radar system further includes a processing circuit, where the processing circuit is configured to perform one or more of the above - described methods (e.g., any one of A1 to A10, B1 to B6, and C1 to C4).

[0087] The above description is an example including one or more embodiments. Of course, it is not possible to describe every conceivable modification and change of the above - mentioned devices or methods for the purpose of describing the above - mentioned methods, but one of ordinary skill in the art can recognize that many additional modifications and permutations of various aspects are possible. Therefore, the described aspects are intended to cover all such changes, modifications, and variations that fall within the spirit and scope of the appended claims. Further, with respect to the term "include" as used in this detailed description or the claims, such term is intended to be inclusive in a manner similar to the way the term "comprising" is interpreted when used as a transitional word in the claims.

Claims

1. An autonomous vehicle AV, comprising: A radar system, the radar system including processing circuitry configured to perform actions, the actions including: Calculating, for a time detection window and based on a frequency modulated continuous wave FMCW signal emitted by the radar system: The speed of a detection in a scene; and The direction of the detection relative to the radar system; Generating a speed - direction histogram for the detection window, where the speed - direction histogram includes bins, and further where the detection is assigned to a bin in the bins based on: The speed assigned to the detection; and The direction of the detection relative to the radar system; Identifying the detection as being due to an interference signal based on the number of detections in the bin of the speed - direction histogram assigned to the detection window; and Generating output data based on the FMCW signal, the output data identifying the detections in the scene, where the detections identified in the output data do not include the detections identified as being due to the interference signal, and where the AV performs driving maneuvers based on the output data.

2. The AV of claim 1, wherein speed and direction are calculated for a number of detections in the time detection window based on the FMCW signal, and wherein the number of detections are assigned to corresponding bins in the speed - direction histogram of the time detection window.

3. The AV of claim 2, wherein the bin in the number of bins has a plurality of detections assigned to it, and further wherein each detection in the plurality of detections assigned to the bin is identified as being due to the interference signal based on the number of detections in the bin of the speed - direction histogram assigned to the detection window.

4. The AV of claim 1, wherein identifying the detection as being due to the interference signal includes: Calculating the number of the detections assigned to the bin in the speed - direction histogram; And Determining that the number of the detections exceeds a threshold, and wherein the detection is identified as being due to the interference signal based on the number of the detections exceeding the threshold.

5. The AV of claim 4, wherein identifying the detection as being due to the interference signal further includes: Calculating the number of the detections assigned to the bin in the speed - direction histogram; Calculating a corresponding number of detections for a bin adjacent to the bin in the speed - direction histogram; Selecting a subset of bins from the bins adjacent to the bin; Determining the number of detections in the subset of bins; And Calculating a difference between the number of the detections assigned to the bin in the speed - direction histogram and the number of detections in the subset of bins, and wherein the detection is identified as being due to the interference signal based on the difference between the number of the detections assigned to the bin and the number of detections in the subset of bins.

6. The AV of claim 5, wherein identifying the detection as being due to the interference signal further includes: Determine that the difference between the number of the detected ones assigned to the bin in the speed - direction histogram and the number of the detected ones in the subset of the bin exceeds a second threshold, wherein the detection is identified as being caused by the interference signal based on the difference between the number of the detected ones assigned to the bin and the number of the detected ones in the subset of the bin exceeding the second threshold.

7. The AV according to claim 5, wherein each detection assigned to the bin is identified as being caused by the interference signal, and wherein the output data does not include each detection assigned to the bin.

8. The AV as claimed in claim 5, wherein the action further comprises: Identify that a second detection is caused by the interference signal, wherein the second detection is assigned to a second bin adjacent to the bin, and wherein identifying that the second detection is caused by the interference signal includes: Select a second subset of bins from the bins adjacent to the bin; Determine the number of detections in the second subset of the bin; Compare the number of detections in the second subset of the bin with the second threshold; and Determine that the number of detections in the second subset of the bin is below the second threshold, wherein the second detection is identified as being caused by the interference signal based on the number of detections in the second subset of the bin being below the second threshold.

9. The AV according to claim 8, wherein each detection adjacent to the bin in each bin is marked as being caused by the interference signal based on the number of detections in the second subset of the bin being below the second threshold.

10. The AV according to claim 1, further comprising a computing system, the computing system communicating with the radar system, wherein the computing system is configured to identify an object in the scene based on the output data.

11. A method performed by a radar system of an autonomous vehicle AV, the method comprising: Receiving signals in a scene; Generating detections based on the received signals and a frequency - modulated continuous - wave FMCW local oscillator within a time detection window, wherein the detections are assigned speed values and direction values; Assigning the detections to bins in a speed - direction histogram for the time detection window based on the speed values and the direction values assigned to the detections, wherein the speed - direction histogram includes a plurality of bins; Determining the number of detections in the bin; Identifying that the signal is emitted from an interference source based on the number of detections in the bin; And After identifying that the signal is emitted from the interference source, generating output data including detections in the time detection window, wherein the detections are not included due to the signal being identified as being emitted from the interference source, and further wherein the AV performs a driving maneuver based on the output data.

12. The method according to claim 11, further comprising: Determining that the number of detections in the bin is greater than a predefined threshold, wherein the signal is identified as being emitted from the interference source based on the number of detections in the bin being greater than the predefined threshold.

13. The method according to claim 11, wherein the bin includes a plurality of detections assigned to the bin, wherein the plurality of detections are assigned the speed value and the direction value, wherein the plurality of detections are identified as being caused by the signal emitted by the interference source, and further wherein the output data does not include the plurality of detections.

14. The method according to claim 11, further comprising: determining, for the time window, the number of detections in a second bin among the plurality of bins, wherein the signal is identified as being emitted by the interference source based on the number of detections in the second bin.

15. The method according to claim 14, further comprising: defining a bin in the speed - direction histogram as a neighboring bin of the bin in the speed - direction histogram; determining the number of detections in a subset of the bins in the neighboring bin; and calculating a difference between the number of detections in the subset of the bin and the number of detections in the bin, wherein the signal is identified as being emitted by the interference source based on the difference between the number of detections in the subset of the bin and the number of detections in the bin.

16. The method according to claim 15, the method further comprising: determining that the difference between the number of detections in the subset of the bin and the number of detections in the bin is greater than a threshold, wherein the signal is identified as being emitted by the interference source based on the difference between the number of detections in the subset of the bin and the number of detections in the bin being greater than the threshold.

17. A radar system, comprising: an antenna configured to detect radar signals; and a processing circuit configured to perform operations, the operations including: calculating, for a time detection window and based on the radar signals: the speed of detections in a scene; and the direction of the detections relative to the radar system; generating a speed - direction histogram for the detection window, wherein the speed - direction histogram includes bins, and further wherein the detections are assigned to bins in the histogram based on: the speed assigned to the detections; and the direction of the detections relative to the radar system; identifying the detections as being caused by an interference source based on the number of detections in the bins in the speed - direction histogram assigned to the detection window; and generating output data based on the radar signals, the output data identifying the detections in the scene, wherein the detections identified in the output data do not include the detections identified as being caused by the interference source.

18. The radar system according to claim 17, wherein an autonomous vehicle is attached to the radar system, and further wherein the autonomous vehicle performs driving maneuvers based on the output data.

19. The radar system according to claim 17, further comprising a transmitter configured to transmit a frequency - modulated continuous - wave (FMCW) signal, wherein the output data is further based on the FMCW signal.

20. The radar system according to claim 19, wherein speed and direction are calculated for a number of detections in the time detection window based on the FMCW signal, wherein the number of detections are assigned to a number of bins in the speed - direction histogram of the detection window, and further wherein each detection assigned to a bin among the number of detections is identified as being caused by the interference source based on the number of detections assigned to the bin in the speed - direction histogram of the detection window.

Images