FMCW ranging device

Abstract

The invention relates to an FMCW ranging device comprising: a light source, in particular a laser, which generates a frequency-modulated transmission beam as a transmission signal with a predetermined frequency deviation and transmits the frequency-modulated transmission beam into a measurement region; a light receiver configured to receive light reflected by an object in the measurement area as a reception signal; a mixer configured to mix at least a portion of the transmit signal with the receive signal to generate a mixed signal; a conversion unit configured to convert the mixed signal into a binary signal; and an evaluation unit configured to determine a distance to the corresponding object based on the binary signal.

Description

Technical Field

[0001] This invention relates to an FMCW ranging device and an FMCW ranging method. Background Technology

[0002] Accurate and reliable ranging plays a central role in modern sensor technology, particularly in applications such as autonomous driving, robotics, industrial automation, and optical measurement. One widely used method for determining object distance is the Frequency-Modulated Continuous Wave (FMCW) method. This method determines object distance by measuring the time delay of the reflected received beam using an optically tuned beam.

[0003] In the FMCW method, the beam is modulated by continuously changing its frequency, and a modulated signal is emitted. A portion of the emitted signal is reflected off the target and received by a photodetector, while another portion is directly conducted to the receiver. Because the reflected signal takes time to reach the target and return, it experiences a phase shift and frequency change upon reaching the receiver. The receiver mixes the directly conducted signal with the reflected signal to generate a periodic signal with a difference frequency, known as a beat signal. The frequency of the beat signal (the beat frequency) is proportional to the distance to the target object, and therefore the distance can be determined by performing a Fourier transform on the electrical signal.

[0004] The FMCW method offers numerous advantages, such as high accuracy, high resolution, and robustness against environmental influences. However, determining the beat frequency typically involves significant effort and places high demands on the corresponding components used for this purpose. One of the biggest obstacles is the need for high-resolution and fast analog-to-digital converters (ADCs) to sample the electrical beat signal with sufficiently high accuracy and speed. These ADCs are not only expensive but also significantly increase system complexity. This is particularly true for applications where multiple channels must be evaluated simultaneously due to the relatively slow measurement speed of the FMCW method. Parallel evaluation of multiple channels requires additional ADCs, which further increases cost and complicates the design of the measurement equipment. Summary of the Invention

[0005] Therefore, the object of the present invention is to provide an improved FMCW ranging device and an improved FMCW ranging method.

[0006] This objective is satisfied by the subject matter of the independent claims.

[0007] The first aspect of the present invention relates to an FMCW ranging device, comprising:

[0008] A light source, particularly a laser, generates a frequency-modulated transmission beam as a transmission signal with a predetermined frequency deviation and transmits the frequency-modulated transmission beam to the measurement area.

[0009] A light receiver, configured to receive light reflected from an object in the measurement area as a received signal;

[0010] A mixer configured to mix at least a portion of a transmitted signal with a received signal to generate a mixed signal;

[0011] A conversion unit configured to convert a mixed signal into a binary signal; and

[0012] The evaluation unit is configured to determine the distance to the corresponding object based on binary signals.

[0013] In other words, the mixed signal generated by the mixer is approximated by a binary signal, which facilitates further data processing. Specifically, the binary signal essentially contains the frequency information of the mixed signal, while the amplitude information can be at least partially discarded. In particular, the binary signal is significantly more compact than the mixed signal, so the frequency of the mixed signal can be determined more quickly with less computational resources.

[0014] As previously described, the light source generates a frequency-modulated transmission beam as a transmission signal with a predetermined frequency deviation and transmits the beam to the measurement area. The transmission beam is reflected by an object in the measurement area, and the reflected light is received as a received signal by a photodetector. Then, a portion of the transmission signal is mixed with the received signal in a mixer to form a mixed signal.

[0015] Frequency deviation can occur, for example, by changing the wavelength of the transmitted light beam, such as rising from a lower frequency (fu) to a higher frequency (fo) within the modulation duration (Tmod) (and vice versa). Therefore, the transmitted signal is particularly a chirped signal. The measurement duration (Tmeas) for determining the mixed frequency of the mixed signal by the ranging device can at most correspond to the modulation duration minus the flight time (Ttof; round trip) until the object is reached at the maximum measurement distance (Tmeas = Tmod - Ttof). The measurement duration is preferably shorter than 10 µs, and particularly preferably shorter than 5 µs.

[0016] Therefore, during the mixing of transmitted and received signals, two signals with different frequencies are mixed. The frequency difference between the transmitted and received signals depends in this respect on the distance between the reflecting object and the ranging device. The greater the distance, the greater the frequency difference. This frequency difference is also designated as the so-called beat frequency.

[0017] During the mixing (or superposition) of transmitted and received signals, a mixed signal, also known as a beat signal, is generated. The beat frequency of the mixed signal corresponds to the frequency difference between the transmitted and received signals.

[0018] The flight time can be calculated from the beat frequency generated by the mixing, and then the distance of the object can be calculated.

[0019] However, determining the beat frequency directly from the mixed signal requires considerable effort. Since the mixed signal is particularly periodic, having values ​​above a threshold in the first half of the period and below a threshold in the second half of the associated period (and vice versa), it can be simplified using a binary signal, where values ​​above the threshold are represented by a binary value of 1 and values ​​below the threshold by a binary value of 0. For example, the start of a period in a binary signal can be represented by a change in the binary signal value from 0 to 1 or from 1 to 0, and the end of the corresponding period can be represented by the next change in the binary signal value from 0 to 1 or from 1 to 0. Based on this simplification, the mixed signal can be converted into a binary signal using a conversion unit. In particular, the binary signal at least approximates the frequency information of the mixed signal. The beat frequency and the distance to the corresponding object can then be estimated or determined based on the binary signal. To further facilitate the determination of the beat frequency, the mixed signal can be amplified using an amplifier (e.g., a differential amplifier). Additionally or alternatively, the mixed signal, the received signal (particularly the electrically converted received signal), and / or the transmitted signal (particularly the electrically converted transmitted signal) can be filtered using appropriate low-pass filters to conform to the sampling theorem. For example, a low-pass filter can be formed by a capacitor in the feedback of a differential amplifier.

[0020] The term "binary signal" in this article refers to a signal that can only assume two possible states and / or values, such as 1 and 0. Binary signals can also assume two different voltage levels, such as "low" and "high".

[0021] The advantage of this invention is that by using binary signals to determine the beat frequency, the requirements for the components used to determine the beat frequency are significantly reduced. In particular, the required computing power is reduced, and the determination of the beat frequency is also simplified. Due to the reduced demand for corresponding components, the cost is further reduced.

[0022] Further embodiments of the invention can be seen in the specification, dependent claims and drawings.

[0023] According to a first embodiment, the conversion unit includes at least one comparator and / or a 1-bit analog-to-digital converter (ADC) configured to convert a mixed signal into a binary signal, wherein the comparator and / or the 1-bit ADC compares the mixed signal with a threshold, wherein if the mixed signal value is greater than the threshold, the binary signal has a predefined first value, wherein if the mixed signal value is less than the threshold, the binary signal has a predefined second value different from the first value, wherein the threshold preferably corresponds to the average value of the mixed signal.

[0024] In other words, each time the beat signal exceeds or falls below a threshold, the output of the comparator and / or the 1-bit ADC changes its state (from low to high or vice versa). Thus, the analog mixed signal can be converted into a binary signal representing a sequence of square wave pulses. The frequencies of these square wave signals specifically correspond to the frequencies of the mixed signal. The quantization resolution is specifically 1 bit. In principle, quantization and sampling rate are interrelated. Therefore, a lower quantization resolution can be compensated for by a higher sampling rate. To provide sufficient signal quality, the ADC's sampling rate can therefore be greater than two, three, four, six, eight, or ten times the maximum beat frequency to be detected.

[0025] The following discussion focuses only on comparators. However, the following statements also apply to 1-bit ADCs.

[0026] The advantage of using a comparator is that it is less susceptible to small amplitude variations in the mixed signal. As long as the mixed signal exceeds the comparator's threshold, a clean square wave signal is generated, which can be easily further processed. Furthermore, comparators can operate very quickly, enabling accurate detection of high frequencies in the beat signal without imposing complex sampling rate requirements on traditional ADCs. It is important to note that the use of a comparator may result in a loss of amplitude information in the original mixed signal, as the comparator essentially transmits the frequency information of the mixed signal. However, this loss of amplitude information is intentionally acceptable, especially since the frequency of the mixed signal (i.e., the beat signal) is related to a defined distance, and this can reduce the requirements for the components used.

[0027] In addition to the above, multiple comparators, particularly two or four, can be used to increase quantization resolution. For example, using two comparators can double the quantization resolution, and using four comparators can quadruple the quantization resolution. Preferably, a maximum of one, two, three, or four comparators are used. To handle additional signals, SERDES blocks (serializer-deserializer blocks) can be further combined with high-speed I / O units to optimize data transmission. Depending on the application, this can thus optimize performance and system cost.

[0028] According to one embodiment, the conversion unit includes a common-mode control unit that determines a threshold based on the mixed signal. The common-mode control unit may, for example, be a circuit that adds the DC components of the mixed signal such that the threshold corresponds to the average value of the mixed signal. For example, the mixed signal may be differentially guided, wherein two lines guiding the mixed signal are connected to the common-mode control unit to add the DC components of the signals guided in these two lines and generate an average value of the added DC components.

[0029] According to one embodiment, the evaluation unit is configured to sample a binary signal at a predefined time step and output the sampled signal as a bitstream, wherein if the sampled value corresponds to a predefined first value, the sampled value in the bitstream is represented as 1, and if the sampled value corresponds to a predefined second value, the sampled value in the bitstream is represented as 0. The bitstream specifically contains frequency information of the mixed signal. Therefore, the beat frequency of the mixed signal can be determined based on the bitstream. In particular, the frequency of the binary signal, and thus the frequency of the mixed signal, i.e., the beat frequency, can be inferred based on the sampling rate or the length of a predefined time step, and with reference to bit changes from 0 to 1 or vice versa (specifically indicating the start or end of a period). Since the signal to be sampled is a binary signal, such as a square wave signal, the signal can also be easily converted to the digital domain.

[0030] According to one embodiment, the conversion unit and / or evaluation unit are configured as part of the FPGA. By reducing the quantization resolution, particularly by abstracting the mixed signal into a binary signal, the beat frequency can be determined in the FPGA with less hardware effort. In particular, a small and therefore cost-effective FPGA can be used. Thus, evaluation can be particularly efficient using an FPGA. Consequently, complete data processing (especially digital data processing) can be performed on the FPGA.

[0031] According to one embodiment, the mixing is performed optically or electrically. In the case of optical mixing, the mixer includes at least one photodiode, which mixes at least a portion of the transmitted signal and the received signal, converting them into an electrically mixed signal. Specifically, the transmitted and received signals are each superimposed in the form of a beam to generate an optically mixed signal, which is then converted into an electrically mixed signal, for example by means of a photodiode. If the mixing of the transmitted and received signals is performed optically, the mixing results in the generation of additive and subtractive mixing frequencies, wherein, in this case, the additive frequencies can each be filtered out by means of a low-pass filter, since these frequencies are not intended for further processing. In this case, the optical receiver and the mixer can particularly be designed as a common component.

[0032] Alternatively, the transmitted signal can be converted into an electrical transmitted signal by means of a photodiode, and the received signal can be converted into an electrical received signal by means of a photodiode, wherein the electrical transmitted signal and the electrical received signal can be mixed (e.g., by addition or multiplication) by means of electrical components (e.g., by analog multiplier, ring modulator or differential amplifier).

[0033] It must be clarified that the mixing in this example should be understood as multiplication and addition or subtraction. In particular, in the case of optical mixing (e.g., for transmitted and received signals), multiplication may be included, while in the case of electrical mixing (e.g., through the aforementioned differential amplifier), addition or subtraction may also be included.

[0034] According to one embodiment, the mixer includes a balanced optical heterodyne detector. For example, in the case of heterodyne detection, at least a portion of the received signal is superimposed on the transmitted signal. Compared to simple detection using only one photodetector, the balanced detector uses, for example, two photodetectors. Specifically, at least a portion of the transmitted signal is superimposed on the received signal, and the superimposed signal is fed to two independent photodetectors. For this purpose, at least a portion of the transmitted signal and the received signal can be separated by means of a beam splitter, such that 50% of at least a portion of the transmitted signal and 50% of the received signal are fed to a first photodetector, and 50% of at least a portion of the transmitted signal and 50% of the received signal are fed to a second photodetector. Specifically, the first photodetector outputs a first mixed signal, while the second photodetector outputs a second mixed signal. Preferably, each of the first and second mixed signals is amplified by means of a transimpedance amplifier, and then the outputs of the two photodetectors (i.e., the first mixed signal and the second mixed signal) are subtracted. The subtraction is preferably performed by means of a differential amplifier. Because of this subtraction, common noise components (such as light source noise or thermal noise) are filtered out, while the actual signal (the difference signal) is amplified. This results in an improvement in the signal-to-noise ratio (SNR) and allows for a more accurate determination of the mixed signal, and thus the mixing frequency.

[0035] According to one embodiment, an optical receiver is configured to convert a received signal into an electrical received signal, wherein another optical receiver is configured to convert at least a portion of a transmitted signal into an electrical transmitted signal, and a mixer includes a differential amplifier configured to mix the electrical transmitted signal and the electrical received signal, outputting the amplified difference between the electrical transmitted signal and the electrical received signal as a mixed signal. Therefore, the mixing of the transmitted and received signals is specifically performed electrically. The differential amplifier can amplify the difference between the electrical transmitted signal and the electrical received signal, thereby simplifying further signal processing. In particular, this allows for higher measurement resolution. In principle, any desired amplifier can be used to amplify the mixed signal, whether for electrical mixing or optical mixing.

[0036] According to one embodiment, a first transimpedance amplifier and a second transimpedance amplifier are connected upstream of a differential amplifier. The first transimpedance amplifier is configured to amplify the electrical transmission signal and feed the amplified electrical transmission signal to the differential amplifier, while the second transimpedance amplifier is configured to amplify the electrical reception signal and feed the amplified electrical reception signal to the differential amplifier. The advantage here is that using two transimpedance amplifiers can reduce the amplifier noise of the differential amplifier.

[0037] It should be noted that even when using a balanced optical heterodyne detector, a differential amplifier can be used to amplify the mixed signal. In this case, the differential amplifier can be configured, for example, such that the electrically mixed signal (i.e., the output signal of the balanced optical heterodyne detector) is connected to the first input and grounded to the other input.

[0038] According to one embodiment, the evaluation unit is configured to determine the frequency of a mixed signal using a bitstream via a neural network, and to determine the distance to a corresponding object based on the determined frequency of the mixed signal. The neural network can be trained with sample data (e.g., sample bit sequences and associated sample frequencies) to determine associated frequencies from the current bitstream. For this purpose, during the training of the neural network, an electrical signal of a known frequency can be converted into a binary signal using a comparator as described above, and then sampled to obtain the corresponding sample bit sequence. The sample bit sequence and its associated frequencies can then be used as training data for training the neural network, where the sample bit sequence serves as input data to the neural network, and the neural network outputs a frequency as an output value, which can be compared with the actual frequency of the sample bit sequence to optimize the various parameters of the neural network.

[0039] According to one embodiment, the evaluation unit is configured to transform the bitstream to the frequency domain, wherein the neural network includes a trained CNN (Convolutional Neural Network) that determines the frequency of a mixed signal based on the transformed bitstream, and the evaluation unit is configured to determine the distance to a corresponding object based on the determined mixed signal frequency. "Transforming the bitstream to the frequency domain" means, for example, interpreting the bitstream as a discrete-time series and transforming it to the frequency domain using a Fast Fourier Transform (FFT). In other words, the frequency components of the bitstream are determined, specifically depending on the transitions between bits (from 0 to 1 or from 1 to 0). For example, a periodic bitstream may have sharp peaks at certain frequencies, while a random (i.e., non-periodic) bitstream may have a wider spectrum. For example, a CNN can be trained to estimate or determine the frequencies associated with signal peaks based on the input data (i.e., the transformed bitstream). The transformed bitstream may specifically have a form, such as a certain number of bits, that adapts to a predefined form of the CNN's input data. For example, a CNN can be configured to process only n-bit (preferably 8-bit) bitstreams. In this scenario, a bitstream transformation is performed so that the transformed bitstream contains exactly n bits. For example, ideally, a bitstream mapping the frequencies of an associated periodic signal can be used for CNN training. Furthermore, the frequencies associated with the bitstream can be used as labels, i.e., as the expected output values ​​of the CNN. By using FFT to transform the bitstream into the transformed bitstream, the CNN can be trained using the transformed bitstream data, where the CNN is optimized based on the deviation between its output values ​​and the expected output values, specifically using the backpropagation algorithm.

[0040] According to one embodiment, a neural network is trained to determine the frequency of a mixed signal based on the bitstream, particularly without transforming the bitstream, wherein an evaluation unit is configured to determine the distance to a corresponding object based on the determined frequency of the mixed signal. For example, this neural network is an anomaly detection neural network. Therefore, the neural network specifically determines the mixing frequency based on the time representation of the bitstream (i.e., not the frequency representation). The neural network specifically determines the mixing frequency directly from the bitstream.

[0041] According to one embodiment, the evaluation unit may include and / or perform a logical AND operation, which in particular results in and / or replaces multiplication.

[0042] According to one embodiment, the neural network may include and / or perform a logical AND operation, which in particular results in and / or replaces multiplication.

[0043] Specifically, if the evaluation unit is configured to perform a transformation to the frequency domain, preferably by transforming the bitstream to the frequency domain, this transformation may include a logical AND operation that results in multiplication. This is achieved by means of 1-bit binaryization of a 1-bit ADC, since when only 1 bit is used, a (single) logical AND operation corresponds to multiplication. Preferably, two 1-bit signals are incorporated into the logical AND operation. In this way, the multiplication typically required for Fourier transform can be mapped to the hardware in a simplified manner.

[0044] Then, neural networks can also be simplified (or simpler solutions can be found during training) because simplified logical operations can be used, for example, 1-bit operations instead of complex multiplications.

[0045] According to one embodiment, the evaluation unit is configured to replicate a binary signal using an observer filter, determine the frequency of a mixed signal based on a comparison between the replicated binary signal and the original binary signal, and determine the distance to the corresponding object based on the determined frequency of the mixed signal. For example, the filter parameters of the observer filter are optimized, particularly continuously, so that the error (i.e., difference) between the original binary signal and the replicated binary signal is minimized. Based on the optimized filter parameters, the frequency of the replicated signal can be determined, and the determined frequency is defined as the frequency of the mixed signal.

[0046] According to one embodiment, the evaluation unit is configured to transform a binary signal to the frequency domain, determine the frequency of a mixed signal based on the transformed signal, and determine the distance to the corresponding object based on the determined frequency of the mixed signal. For example, the binary signal can be transformed to the frequency domain using a Fast Fourier Transform (FFT), and the frequency can be determined from the FFT of the binary signal using known analytical methods. Additionally or alternatively, the evaluation unit can be configured to transform a bitstream to the frequency domain and determine the frequency of the mixed signal from the transformed bitstream using known analytical methods. Furthermore, it is advantageous to evaluate the signal peak values ​​of the harmonics of the output signal in the frequency domain. This improves the accuracy of determining the fundamental frequency (i.e., the beat frequency) and thus ultimately further improves the signal-to-noise ratio.

[0047] In a preferred embodiment, the FMCW ranging device may preferably include multiple measurement channels, each of which in each case has a light source, a light receiver, and a mixer. A common conversion unit may be configured to convert the mixed signal of the respective measurement channel into a binary signal. A common evaluation unit may be configured to determine the distance to the respective object based on the binary signal. Preferably, the common conversion unit and / or the common evaluation unit may be configured as part of a single FPGA. The mixed signal of the respective measurement channel is further preferably fed to a SERDES block of the FPGA and, in particular, converted into a binary signal by the SERDES block respectively. This allows for the advantageous implementation of a parallel system. In particular, due to the 1-bit conversion, hardware savings are possible, enabling the implementation of a multi-channel system within a single FPGA.

[0048] Another aspect of the present invention relates to an FMCW ranging method, wherein:

[0049] A frequency-modulated transmission beam is generated as a transmission signal and transmitted to the measurement area, wherein the transmission signal has a predetermined frequency deviation.

[0050] The light reflected from objects in the measurement area is used as the received signal.

[0051] At least a portion of the transmitted signal is mixed with the received signal to generate a mixed signal.

[0052] Convert the mixed signal into a binary signal.

[0053] The distance to the corresponding object is determined based on the binary signal.

[0054] The statements regarding the FMCW ranging device according to the invention apply accordingly to the method; this is particularly true regarding the advantages and embodiments.

[0055] It should be noted that any combination of the above embodiments is possible unless explicitly excluded. Attached Figure Description

[0056] The invention will now be presented by way of example only with reference to the accompanying drawings. The following shows:

[0057] Figure 1 This is a schematic diagram of the FMCW ranging device;

[0058] Figure 2 It is a component of the data processing unit used to determine the distance to an object;

[0059] Figure 3 This is a schematic diagram of the FMCW ranging device;

[0060] Figure 4The FPGA of an FMCW ranging device that uses a CNN to determine the beat frequency is schematically shown;

[0061] Figure 5 The FPGA of an FMCW ranging device that uses an anomaly detection neural network to determine the beat frequency is schematically illustrated; and

[0062] Figure 6 This is a schematic diagram of the operating modes of the observer filter. Detailed Implementation

[0063] Figure 1 A schematic diagram of an FMCW ranging device 12 is shown. The FMCW ranging device includes a laser 14 that generates a frequency-modulated transmission beam as a transmission signal 16 with a predetermined frequency deviation and transmits the frequency-modulated transmission beam to the measurement area. The FMCW ranging device 12 also includes a light receiver 18 that receives light reflected from an object 20 in the measurement area as a received signal 22. The received signal 22 can be fed to the light receiver 18, in particular by means of a circulator (not shown), for example, the transmission light is radiated via the circulator, and the received light is picked up and forwarded. A mixer 24 of the FMCW ranging device 12 mixes at least a portion of the transmission signal 16 with the received signal 22 to generate a mixed signal. A conversion unit 26 then converts the mixed signal into a binary signal, and an evaluation unit 28 determines the distance to the object 20 based on the binary signal. The conversion unit 26 and the evaluation unit 28 are configured as part of a data processing unit 30 (e.g., an FPGA 31).

[0064] Figure 2The diagram illustrates components of a data processing unit 30 for determining the distance to object 20. The data processing unit 30 includes a differential amplifier 32 that generates and / or amplifies a mixed signal. For example, the mixed signal, previously converted to an electrical signal, can be differentially guided, where two differentially guided portions of the mixed signal are applied to the input of the differential amplifier 32 to generate an amplified mixed signal. Alternatively, a first photodiode can convert at least a portion of the transmitted signal 16 into an electrical transmitted signal, while a second photodiode can convert the received signal 22 into an electrical received signal, where the electrical transmitted and received signals are applied to their respective inputs to the differential amplifier 32 to amplify the difference between the electrical transmitted and received signals, thus generating a mixed signal. The mixed signal is then filtered by a low-pass filter 34 to conform to the sampling theorem. Based on the amplified, filtered mixed signal, a common-mode control circuit 36 ​​can determine a threshold for the comparator 38, preferably corresponding to the DC component of the mixed signal, i.e., the value around the center line around which the periodic mixed signal oscillates. Based on the threshold and the mixed signal, comparator 38 generates a binary signal, such as a rectangular voltage signal. If the mixed signal is greater than the threshold, the value is assumed to be "high," and if the mixed signal is less than the threshold, the value is assumed to be "low." Then, evaluation unit 28 (not shown) uses the binary signal to calculate the distance to object 20.

[0065] Figure 3An embodiment of the FMCW ranging device 12 is shown, in which two photodetectors 42, 44 are used to generate a mixed signal or a striking signal. In this respect, at least a portion of the transmitted signal 16 and the received signal 22 are superimposed by two independent photodetectors 42, 44. For this purpose, at least a portion of the transmitted signal 16 and the received signal 22 can be divided separately by a beam splitter 40, such that 50% of the transmitted signal 16 and 50% of the received signal 22 are fed to the first photodetector 42, and 50% of the transmitted signal 16 and 50% of the received signal 22 are fed to the second photodetector 44. The first photodetector 42 outputs a first mixed signal, while the second photodetector 44 outputs a second mixed signal. The first mixed signal is then amplified by a first transimpedance amplifier 43, and the second mixed signal is amplified by a second transimpedance amplifier 45. The amplified first mixed signal and the amplified second mixed signal are then subtracted to produce the final mixed signal. The subtraction is performed by a differential amplifier 32. Due to the subtraction, common noise components (such as light source noise or thermal noise) are filtered out, while the actual signal (difference signal) is amplified. This results in an improved signal-to-noise ratio and allows for more accurate determination of the mixed signal and mixing frequency. Finally, the final mixed signal is transferred to FPGA 31, which converts it into a binary signal and determines the mixing frequency (i.e., beat frequency) of the final mixed signal based on this binary signal. The distance to object 20 is then determined or calculated based on the beat frequency.

[0066] Figure 4 An FPGA 31 is schematically illustrated for determining the beat frequency using a CNN 50. As described above, the FPGA 31 is connected to an analog preprocessing circuit 46, which includes, for example, a laser 14, a light receiver 18, and / or a mixer 24. The FPGA 31, particularly the comparator 38 arranged on the FPGA 31, receives the mixed signal and converts it into a binary signal, then samples it and outputs it as an n-bit stream. An FFT 48 is then performed based on the bit stream, with the result of the FFT 48 output as a transformed bit stream. Using the transformed bit stream, the frequency of the mixed signal is determined by the CNN 50, wherein the distance calculation unit 52 determines the distance to the corresponding object 20 based on the determined frequency of the mixed signal.

[0067] Figure 5 The FPGA 31 used to determine the beat frequency by means of an anomaly detection network 54 is schematically illustrated. Figure 4 Instead, the bitstream is not used for FFT 48. Instead, the beat frequency or distance to the corresponding object 20 is determined directly or immediately based on the bitstream determined from the binary signal by means of the anomaly detection neural network 54.

[0068] Figure 6 A schematic diagram of the operating mode of observer filter 56 is shown. Observer filter 56 is configured to replicate, for example, a binary signal output by comparator 38. The filter parameters of observer filter 56 are continuously optimized in this regard so that the error (i.e., difference) between the binary signal and the replicated binary signal is minimized. Then, distance calculation unit 52 can determine the frequency of the replicated signal based on the optimized filter parameters, and can define the determined frequency of the replicated signal as the frequency of the mixed signal, i.e., the beat frequency.

[0069] List of reference numerals

[0070] 12 FMCW ranging device

[0071] 14 Lasers

[0072] 16. Transmitted Signals

[0073] 18 Optical Receivers

[0074] 20 objects

[0075] 22 Received signal

[0076] 24 Mixer

[0077] 26 Conversion Units

[0078] 28 Evaluation Units

[0079] 30 Data Processing Units

[0080] 31 FPGA

[0081] 32 Differential Amplifier

[0082] 34 Low-pass filter

[0083] 36 Common-mode control circuit

[0084] 38 comparators

[0085] 40-beam beam splitter

[0086] 42 First photodetector

[0087] 43 First Transimpedance Amplifier

[0088] 44 Second photodetector

[0089] 45 Second Transimpedance Amplifier

[0090] 46 Analog Preprocessing Circuit

[0091] 48 FFT

[0092] 50 CNN

[0093] 52 Distance Calculation Unit

[0094] 54 Anomaly Detection Neural Network

[0095] 56 Observer Filter

Claims

1. An FMCW ranging device, comprising: A light source generates a frequency-modulated transmission beam as a transmission signal with a predetermined frequency deviation, and transmits the frequency-modulated transmission beam to the measurement area. A light receiver is configured to receive light reflected from an object in the measurement area as a received signal; A mixer configured to mix at least a portion of the transmitted signal with the received signal to generate a mixed signal; A conversion unit configured to convert the mixed signal into a binary signal; and An evaluation unit is configured to determine the distance to the corresponding object based on the binary signal.

2. The FMCW ranging device according to claim 1, characterized in that, The light source is a laser.

3. The FMCW ranging device according to claim 1, characterized in that, The conversion unit includes at least one comparator or a 1-bit analog-to-digital converter.

4. The FMCW ranging device according to claim 1, characterized in that, The conversion unit is configured to convert the mixed signal into a binary signal, wherein a comparator or a 1-bit analog-to-digital converter compares the mixed signal with a threshold; if the mixed signal value is greater than the threshold, the binary signal has a predefined first value, and if the mixed signal value is less than the threshold, the binary signal has a predefined second value different from the first value.

5. The FMCW ranging device according to claim 4, characterized in that, The threshold corresponds to the average value of the mixed signal.

6. The FMCW ranging device according to claim 1, characterized in that, The conversion unit includes a common-mode control unit, which determines a threshold based on the mixed signal.

7. The FMCW ranging device according to claim 3, characterized in that, The evaluation unit is configured to sample the binary signal at a predefined time step and output the sampled signal as a bit stream; if the sampled value approximately corresponds to a predefined first value, the sampled value in the bit stream is represented as 1, and if the sampled value approximately corresponds to a predefined second value, the sampled value in the bit stream is represented as 0.

8. The FMCW ranging device according to claim 6, characterized in that, The evaluation unit is configured to sample the binary signal at a predefined time step and output the sampled signal as a bit stream; if the sampled value approximately corresponds to a predefined first value, the sampled value in the bit stream is represented as 1, and if the sampled value approximately corresponds to a predefined second value, the sampled value in the bit stream is represented as 0.

9. The FMCW ranging device according to claim 1, characterized in that, The conversion unit and / or the evaluation unit are configured as part of the FPGA.

10. The FMCW ranging device according to claim 1, characterized in that, Signal mixing is performed in an optical or electrical manner.

11. The FMCW ranging device according to claim 10, characterized in that, The mixer includes a balanced optical heterodyne detector.

12. The FMCW ranging device according to claim 10, characterized in that, The optical receiver is configured to convert the received signal into an electrical received signal. Another optical receiver is configured to convert at least a portion of the transmitted signal into an electrical transmission signal, and The mixer includes a differential amplifier configured to mix the electrical transmission signal and the electrical reception signal, and output the difference between the amplified electrical transmission signal and the electrical reception signal as a mixed signal.

13. The FMCW ranging device according to claim 12, characterized in that, The first and second transimpedance amplifiers are connected upstream of the differential amplifier. The first transimpedance amplifier is configured to amplify the electrical transmission signal and feed the amplified electrical transmission signal to the differential amplifier. The second transimpedance amplifier is configured to amplify the electrical received signal and feed the amplified electrical received signal to the differential amplifier.

14. The FMCW ranging device according to claim 1, characterized in that, The evaluation unit is configured to determine the frequency of the mixed signal using a bitstream via a neural network, and to determine the distance to the corresponding object based on the determined frequency of the mixed signal.

15. The FMCW ranging device according to claim 14, characterized in that, The evaluation unit is configured to transform the bitstream to the frequency domain; the neural network includes a trained CNN that determines the frequency of the mixed signal based on the transformed bitstream. The evaluation unit is configured to determine the distance to the corresponding object based on the determined frequency of the mixed signal.

16. The FMCW ranging device according to claim 14, characterized in that, The neural network is trained to determine the frequency of the mixed signal based on the bitstream. The evaluation unit is configured to determine the distance to the corresponding object based on the determined frequency of the mixed signal.

17. The FMCW ranging device according to claim 16, characterized in that, The neural network is trained to determine the frequency of the mixed signal based on the bitstream without transforming the bitstream.

18. The FMCW ranging device according to claim 1, characterized in that, The evaluation unit is configured to replicate the binary signal using an observer filter, determine the frequency of the mixed signal based on a comparison between the replicated binary signal and the original binary signal, and determine the distance to the corresponding object based on the determined frequency of the mixed signal.

19. The FMCW ranging device according to claim 1, characterized in that, The evaluation unit is configured to transform the binary signal to the frequency domain, determine the frequency of the mixed signal based on the transformed signal, and determine the distance to the corresponding object based on the determined frequency of the mixed signal.

20. An FMCW ranging method, wherein: A frequency-modulated transmission beam is generated as a transmission signal, and the frequency-modulated transmission beam is transmitted to the measurement area, wherein the transmission signal has a predetermined frequency deviation. The light reflected from the object in the measurement area is received as a signal. At least a portion of the transmitted signal is mixed with the received signal to generate a mixed signal. Convert the mixed signal into a binary signal, and The distance to the corresponding object is determined based on the binary signal.

Images