Apparatus, radar system, and method for estimating heart rate

Abstract

The invention relates to a device, a radar system and a method for estimating heart rate. The present disclosure relates to an apparatus for measuring a heart rate of a target in a field of view of a radar sensor. The apparatus includes processing circuitry configured to determine a frequency spectrum corresponding to the time domain data based on the received signal. The processing circuitry is further configured to determine a respiratory rate frequency based on the maximum amplitude of the spectrum. Further, the processing circuitry is configured to determine a plurality of segment spectra based on the respiratory rate frequency, each segment spectrum corresponding to a respective segment of the time domain data, and to determine an average of the segment spectra to obtain an average spectrum. Further, the processing circuitry is configured to subtract the average spectrum from the spectrum to obtain a differential spectrum. Based on the maximum amplitude of the differential spectrum, the processing circuitry is configured to estimate the heart rate of the target.

Description

Technical Field

[0001] The present disclosure relates to radar data processing. In particular, examples relate to an apparatus, radar system, method, and program for estimating heart rate. Background Art

[0002] Vital signs such as respiratory rate and heart rate are physical measurements of the body that can be used to monitor the overall health of an organism. Although respiratory rate and heart rate can be measured by various methods, these methods may not be feasible in some cases. In particular, for most stationary targets, high-speed resolution for detecting respiratory rate can be more easily obtained compared to the more challenging heart rate. Also for radar systems, the detection of heart rate faces many challenges due to environmental noise, harmonic components of respiratory rate, and random body movement. Therefore, there is a need to improve radar data processing for heart rate detection. Summary of the invention

[0003] This need is met by the subject matter of the independent claims. Further advantageous embodiments are given by the dependent claims.

[0004] According to a first aspect, the present disclosure relates to an apparatus for measuring the heart rate of a target in a field of view of a radar sensor. The apparatus includes a processing circuit device configured to determine a frequency spectrum corresponding to time domain data based on a received signal. The processing circuit device is also configured to determine a respiration rate frequency based on a maximum amplitude of the frequency spectrum. In addition, the processing circuit device is configured to determine a plurality of segment frequency spectra based on the respiration rate frequency, each segment frequency spectrum corresponding to a respective segment of the time domain data, and is configured to determine a mean of the segment frequency spectra to obtain an average spectrum. The processing circuit device is also configured to subtract the average spectrum from the spectrum to obtain a differential spectrum, and is configured to estimate the heart rate of the target based on the maximum amplitude of the differential spectrum.

[0005] According to a second aspect, the present disclosure relates to a radar system comprising an apparatus as described herein and a radar sensor configured to generate a receive signal.

[0006] According to a third aspect, the present disclosure relates to a method for measuring the heart rate of a target in the field of view of a radar sensor. The method includes determining a frequency spectrum corresponding to time domain data based on a received signal. The method also includes determining a respiration rate frequency based on a maximum amplitude of the frequency spectrum. In addition, the method includes determining a plurality of segment frequency spectra based on the respiration rate frequency, each segment frequency spectrum corresponding to a respective segment of the time domain data, and also includes determining a mean of the segment frequency spectra to obtain an average spectrum. In addition, the method includes subtracting the average spectrum from the frequency spectrum to obtain a differential spectrum, and estimating the heart rate of the target based on the maximum amplitude of the differential spectrum.

[0007] According to a fourth aspect, the present disclosure relates to a program having a program code for performing a method as described herein, when the program is executed on a processor or programmable hardware. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Some examples of apparatus and / or methods will be described below, by way of example only, and with reference to the accompanying drawings, in which

[0009] Figure 1 schematically illustrates an exemplary apparatus for measuring the heart rate of a target in a field of view of a radar sensor;

[0010] Figure 2A An exemplary spectrum is depicted;

[0011] Figure 2B An exemplary average spectrum is depicted;

[0012] Figure 3 A flow chart of exemplary pre-processing steps that may be applied prior to determining a spectrum is provided;

[0013] Figure 4 A visual representation of exemplary processing steps for measuring the heart rate of a target in a field of view of a radar sensor is provided;

[0014] Figure 5 depicts a range-time curve corresponding to time-domain data within a specified range bin of a range Fourier transform;

[0015] Figure 6 A flow chart of exemplary processing steps that may be applied to measure the heart rate of a target in a field of view of a radar sensor is provided;

[0016] Fig. 7A depicts an exemplary spectrum and an exemplary average spectrum plotted together;

[0017] Figure 7B depicts an exemplary windowed spectrum for a corresponding correlation operation;

[0018] Figure 7C depicts an exemplary correlation spectrum and an exemplary correlation average spectrum generated by corresponding correlation operations;

[0019] Figure 8 provides a zoomed-in depiction of a heart rate component of an exemplary difference spectrum taken from an exemplary correlation spectrum;

[0020] Fig. 9 schematically illustrates an exemplary radar system for measuring the heart rate of a target in a field of view of a radar sensor; and

[0021] Fig.10A flow chart of an exemplary method for measuring the heart rate of a target in a field of view of a radar sensor is provided. DETAILED DESCRIPTION

[0022] Some examples are now described in more detail with reference to the accompanying drawings. However, other possible examples are not limited to the features of these embodiments described in detail. Other examples may include modifications of features and equivalents and substitutes of features. In addition, the terms used to describe some examples herein should not limit other possible examples.

[0023] Throughout the description of the drawings, the same or similar reference numerals refer to the same or similar elements and / or features, which may be identical or implemented in a modified form while providing the same or similar functions. For clarity, the thickness of the lines, layers and / or regions in the drawings may also be exaggerated.

[0024] When two elements A and B are combined using "or", this should be understood to disclose all possible combinations, i.e. only A, only B, and A and B, unless otherwise explicitly defined in individual cases. As an alternative wording for the same combination, "at least one of A and B" or "A and / or B" can be used. The same applies to combinations of more than two elements.

[0025] If singular forms such as "a", "an", and "the" are used, and there is no explicit or implicit definition that the use of only a single element is mandatory, other examples may also use several elements to achieve the same function. If a function is described below as being implemented using multiple elements, other examples may use a single element or a single processing entity to achieve the same function. It should also be understood that the terms "comprise", "comprising", "including", and / or "comprising" when used describe the presence of specific features, wholes, steps, operations, processes, elements, components, and / or groups thereof, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, processes, elements, components, and / or groups thereof.

[0026] Figure 1 1 is a block diagram of an example of an apparatus 100. Consider the apparatus 100 in the context of a radar sensor. For example, the apparatus 100 may be integrated into a radar system including a radar sensor, as described below with reference to Fig. 9 As explained, or can be outside the radar system. In the former case, the device 100 can be outside the radar sensor or integrated (e.g., partially or completely) into the radar sensor. For example, the device 100 can be distributed between the radar sensor and a location outside the radar sensor.

[0027] The apparatus 100 includes a processing circuit device 110 and optionally includes an interface circuit device 120. Where the interface circuit device 120 is present, the interface circuit device 120 may be communicatively coupled (e.g., via a wired or wireless connection) to the processing circuit device 110, e.g., for data exchange between the interface circuit device 120 and the processing circuit device 110. The interface circuit device 120 may be any device or device for transmitting or exchanging data. Where the apparatus 100 includes the interface circuit device 120, the interface circuit device 120 may be configured to receive data indicative of a received signal of a radar sensor. For example, the interface circuit device 120 may be communicatively coupled to a radar sensor or a storage device storing data. The interface circuit device 120 may receive data, e.g., via a wired or wireless coupling to a radar sensor or a storage device.

[0028] Depending on the specific implementation, the device 100 can omit the interface circuit device 120. For example, the processing circuit device 110 can determine the data. For example, the processing circuit device 110 can be integrated into a radar sensor. The radar sensor can be any device that uses radio waves to, for example, detect and locate objects. The radar sensor can be, for example, an FMCW (frequency modulated continuous wave) radar sensor. For example, the radar sensor can be configured to transmit a radio frequency signal (Tx signal) into the field of view (scene) of the radar sensor through a transmitter, and receive reflections (echoes; Rx signals) of the radar frequency signal through a receiver. The radar sensor or an external device coupled to the radar sensor can generate radar data based on the reflections of the received radio frequency signal by, for example, sampling the received reflections with the aid of an analog-to-digital converter (ADC).

[0029] The radio frequency signal (i.e., the received signal) may be in the form of a plurality of chirps, including a portion of the received reflected signal associated with the corresponding transmitted chirp. The chirp may be a radio frequency signal whose frequency varies with time. The frequency of the chirp may sweep over a specific frequency range, such as a chirp bandwidth. For example, the chirp may be a linearly modulated signal, i.e., a signal whose frequency increases or decreases linearly with time.

[0030] The received signal may be any type of signal generated in a receiver of a radar sensor. The received signal may be an intermediate frequency IF signal, which may be generated, for example, by mixing a reflected signal with a local oscillator signal at a specific frequency. The processing circuit device 110 may determine data indicating the received signal by, for example, sampling the received signal, and perform further processing of the data within the radar sensor. The processing circuit device 110 may optionally modify the sampled received signal in a pre-processing step, for example, for noise reduction, removal of DC (direct current), etc. For example, the device 100 may include a memory configured to store such generated data.

[0031] The radar sensor can use a mixer to mix the received echo with a copy of the transmitted signal to produce an IF signal x IF (t) (e.g., a beat signal). The radar sensor may include one or more antennas to receive the reflected signal, and the radar sensor may include an amplifier to receive the reflected signal from its antenna. The beat signal x IF (t) may be filtered using a low pass filter (LPF) and then sampled by an ADC. The ADC may advantageously be capable of sampling the filtered beat signal x at a sampling frequency that is less than the frequency of the received signal. out (t) Sampling is performed, and the received signal is received by the receiving antenna.

[0032] Alternatively, processing circuit device 110 may partially determine the data. For example, processing circuit device 110 may determine a first portion of the data, and at least one external processing circuit device may determine at least a second portion of the data. Processing circuit device 110 and external processing circuit device may be connected, for example, within a distributed computing environment, for jointly determining the data. In this case, processing circuit device 110 may be integrated into the radar sensor, or may be external to the radar sensor. Processing circuit device 110 may receive the second portion of the data, for example, via an interface to an external processing circuit device such as interface circuit device 120, and further process the first portion and the second portion of the data.

[0033] In another alternative, processing circuit device 110 is partially integrated into the radar sensor and partially outside the radar sensor. In this case, interface circuit device 120 is optional. Processing circuit device 110 can, for example, include a first part (first processing circuit device) integrated into the radar sensor and a second part (second processing circuit device) outside the radar sensor. In this case, the determination and / or further processing of the data can be performed in a distributed manner by the first part and the second part of processing circuit device 110.

[0034] Processing circuit device 110 can be, for example, a single dedicated processor, a single shared processor, or multiple individual processors (some or all of which can be shared), digital signal processor (DSP) hardware, application specific integrated circuit (ASIC), microcontroller or field programmable gate array (FPGA). Processing circuit device 110 can be optionally coupled to read-only memory (ROM), random access memory (RAM) and / or non-volatile memory, for example, for storing software.

[0035] As part of the above-described radar system and its multiple components, various embodiments of the apparatus 100 described herein are provided for determining the heart rate and respiration rate of a target in the field of view of the radar sensor. To this end, the apparatus 100 is configured to receive data based on a received signal, the data including time domain data 130, and perform processing steps based on the time domain data 130. The time domain data 130 can be inspected and / or processed into other forms of data before being used to generate a plurality of spectra that can be analyzed for determining the heart rate and respiration rate of the target.

[0036] The processing circuitry 110 of the apparatus 100 is configured to determine a frequency spectrum corresponding to the time domain data 130, and is configured to determine a respiration rate frequency based on a maximum amplitude therein. The respiration rate frequency may be determined by analysis of the frequency spectrum, including searching for and identifying a maximum amplitude of the frequency spectrum. The respiration rate frequency may be determined by identifying a frequency bin within which the amplitude is at a maximum value for the entire frequency spectrum. The respiration rate frequency may be set to a value within a frequency range corresponding to the frequency bin, which may be equivalent to a center frequency within the frequency bin.

[0037] The spectrum can take the form of various spectrum types. For example, the spectrum can be an amplitude spectrum, where the spectrum includes multiple amplitudes, each amplitude corresponding to a corresponding frequency bin (or frequency component). In a specific example, the spectrum can be an amplitude spectrum representing the amplitude or absolute value of the complex amplitude in the frequency domain. Therefore, the amplitude spectrum can provide information about the intensity or strength corresponding to the frequency component without considering the phase information. Alternatively, the spectrum can be a power spectrum representing the power of each frequency component. The spectrum can also be a normalized spectrum, which represents the amplitude or power of the frequency component scaled to a specific reference value or normalized to a specific value. The various types of spectra presented for the above-mentioned spectra can also be similarly applied to all other spectra used in further processing.

[0038] The processing circuit device 110 is also configured to determine a plurality of segment spectra, each segment spectrum corresponding to a respective segment of the time domain data 130. In addition, the plurality of segment spectra are determined based on the breathing rate frequency. For example, the time domain data 130 may be divided into a plurality of segments, wherein each segment corresponds to a time interval, which may be the same amount of time for each segment. The time interval may be defined by the inverse of the breathing rate frequency. For example, the inverse of the breathing rate frequency may provide a time interval equivalent to the amount of time between consecutive breaths, which are performed at a frequency equivalent to the breathing rate frequency. Therefore, each segment of the time domain data 130 may include single breaths at similar time instances within the time interval.

[0039] The processing circuit device 110 is also configured to determine the mean of the segment spectrum to obtain an average spectrum. Each segment spectrum in the segment spectrum may include a common frequency range and a common number of frequency bins (e.g., the same resolution). The processing circuit device 110 may be configured to sum the corresponding amplitudes of each segment spectrum and divide the sum by the number of segment spectra. More specifically, the corresponding amplitude values ​​of each segment spectrum corresponding to a specific frequency bin (such as an equivalent position or indexed frequency bin) may be summed and divided by the number of segment spectra. This may also be done for each frequency bin of the spectrum so that each frequency bin has an average amplitude value associated with it. Multiple average amplitude values ​​associated with each corresponding frequency bin may be used to determine the average spectrum.

[0040] The processing circuit device 110 is configured to subtract the average spectrum from the (original) spectrum to obtain a difference spectrum. For the subtraction, the corresponding amplitude value of each frequency bin of the average spectrum can be subtracted from the corresponding amplitude value of each frequency bin of the (original) spectrum. For example, the amplitude value of a specific frequency bin of the average spectrum can be subtracted from the amplitude value of a similar frequency bin with an equivalent position or index of the (original) spectrum. This can be done for each frequency bin of the average spectrum and the (original) spectrum to obtain a difference for each frequency bin. The corresponding difference values ​​can then be used to obtain a complete difference spectrum. The difference spectrum can also be considered as a difference data set with corresponding amplitude difference values, where the difference data set can be plotted according to the corresponding frequency bins to provide a difference (frequency) spectrum.

[0041] In addition, the processing circuit device 110 is configured to estimate the heart rate of the target based on the maximum amplitude of the differential spectrum. The heart rate may correspond to a specific heart rate frequency, which may correspond to the maximum amplitude of the differential spectrum. Similar to finding the respiratory rate frequency based on the maximum amplitude, the heart rate frequency may be determined by analyzing the differential spectrum, including searching for and identifying the maximum amplitude of the differential spectrum. The heart rate frequency may be determined by identifying a frequency bin in which the amplitude is at a maximum value for the entire differential spectrum. The heart rate frequency may be determined as a value within a frequency range that corresponds to the frequency bin, which may be equivalent to the center frequency within the frequency bin.

[0042] Therefore, the processing circuit device 110 is configured to obtain a heart rate frequency (e.g., an estimated heart rate) and obtain (i.e., determine) a respiratory rate frequency. The heart rate frequency and the respiratory rate frequency can each be used to provide health information related to a target in the field of view of the radar sensor. Such heart rate frequency and respiratory rate frequency can also be recorded multiple times, which can be performed over a period of days, weeks, months, or years to obtain long-term health information related to the target. The device 100 may include other features that can improve the accuracy of the respiratory rate frequency and / or heart rate frequency obtained. These features will refer to Figures 3 to 8The features of the device 100 previously discussed are also discussed in more detail in the accompanying drawings. Figure 2A and Figure 2B illustrative examples are discussed in more depth.

[0043] Figure 2A An exemplary spectrum 200-1 is depicted. The y-axis may correspond to the amplitude of the frequency. As previously described, the amplitude may correspond to the magnitude or power of the spectrum, which may be normalized or rescaled relative to a particular value. The x-axis of spectrum 200-1 may correspond to the value of the frequency within a pre-specified frequency range. The frequency of the x-axis may be measured in units of minutes, which may be any arbitrary unit per minute (e.g., breaths per minute or heart beats per minute). The depicted minutes are proportional to a scale measured in units of Hertz (Hz), such as s. -1 Or a fraction of a second.

[0044] Spectrum 200-1 is shown as having a first amplitude peak 210-1 corresponding to a maximum amplitude. Maximum amplitude peak 210-1 is located at a first frequency, which may correspond to a respiratory rate frequency. For example, the respiratory rate frequency may be 0.75 Hz, which would correspond to 45 breaths per 60 seconds, or 45 breaths per minute, which is a respiratory rate within the typical range for newborns (typically ranging from 30 to 60 breaths per minute). The maximum amplitude is depicted as corresponding to a value of 45 on the x-axis, or 45 breaths per minute.

[0045] The spectrum 200-1 is also shown as having a second amplitude peak 210-2 at a second frequency that is twice the frequency of the breathing rate. It can be seen that the second frequency corresponds to 90 breaths per minute (1.5 Hz). Therefore, the second amplitude peak 210-2 can also be referred to as the first harmonic of the breathing rate frequency. In addition, the spectrum 200-1 is shown as having a third amplitude peak 210-3 at a third frequency that is three times the frequency of the breathing rate. It can be seen that the third frequency corresponds to 135 breaths per minute (2.25 Hz). Therefore, the third amplitude peak 210-3 can also be referred to as the second harmonic of the breathing rate frequency. The use of the term "amplitude peak" refers to a local maximum that is significantly larger than other possible local maxima, so as to distinguish itself from other local maxima whose amplitudes are not significantly larger than various local minima of the same spectrum.

[0046] Additionally, spectrum 200-1 includes a fourth peak 212 corresponding to a fourth frequency slightly below the third introduced frequency. The fourth peak may correspond to a heart rate. Within the depicted scale, the fourth peak may correspond to a heart rate of approximately 125 beats per minute (approximately 2.08 Hz), slightly below the second harmonic frequency (2.25 Hz) of 135 breaths per minute. A heart rate of 125 beats per minute is a heart rate within the usual range of a newborn's resting heart rate (typically ranging from 100 to 160 beats per minute).

[0047] Figure 2B An exemplary average spectrum 200-2 is depicted. As previously described, the average spectrum 200-2 is based on the breathing rate frequency. The time domain data 130 can be divided into a plurality of segments having a common time interval defined by the breathing rate frequency. For example, the common time interval can be the inverse of the breathing rate frequency, such as the previously mentioned breathing rate frequency of 45 breaths per minute or 0.75 Hz, as shown in FIG. Figure 2A As depicted. For this example, the time interval may be approximately 1.33 seconds in time. More specifically, the time domain data 130 may be divided into a plurality of non-overlapping segments, wherein each segment includes the time domain data 130 including data recorded within a continuous time interval, each time interval may be equivalent to an amount of time of 1.33 seconds. Then, each of the segments corresponding to a portion of the time domain data 130 within a corresponding window of 1.33 seconds may determine a corresponding segment spectrum.

[0048] The average of the multiple segment spectra can be calculated as described above to obtain the average spectrum 200-2, such as Figure 2B Depicted. The average spectrum 200-2 also includes a first amplitude peak 220-1 that is similar to the first amplitude peak 210-1 of the (original) spectrum 200-1, which has similar characteristics. For example, the first amplitude peak 220-1 is also centered on the respiratory rate frequency, which is also approximately 45 breaths per minute (0.75 Hz). The average spectrum 200-2 also includes a second amplitude peak 220-2 that is similar to the second amplitude peak 210-2 of the (original) spectrum 200-1, which has similar similar characteristics. As depicted, this frequency is also located at the first harmonic of the respiratory rate frequency. The same situation is also depicted for the third amplitude peak 220-3 that is similar to the third amplitude peak 210-3 of the (original) spectrum 200-1, which is also located at the second harmonic of the respiratory rate frequency.

[0049] The main difference between the depicted (original) spectrum 200-1 and the average spectrum 200-2 is the lack of a heart rate component. More precisely, a certain frequency component in the average spectrum 200-2, which is close to the heart rate frequency of 125 beats per minute (about 2.08 Hz), consists only of noise. In the average spectrum 200-2, this frequency is depicted with an arrow and a label showing that "the heart rate component has been filtered". When a time interval corresponding to the target respiration rate (i.e., the inverse of the respiration rate frequency) is selected for each of the segment spectra, the heart rate can be effectively filtered or at least mostly filtered.

[0050] In this scenario, each segment of the time domain data 130 at similar locations of the corresponding segments will have amplitude values ​​corresponding to the target's inhalation and exhalation. Therefore, the breathing rate frequency will be depicted very similarly in the average spectrum 200-2 compared to the (original) spectrum 200-1. In addition, the harmonic components are hardly affected. When the time domain data 130 is divided into segments corresponding to the inverse of the fundamental frequency, each segment captures a complete cycle of the signal. In other words, by dividing the time domain data 130 according to the fundamental frequency, each segment is aligned with a specific phase of the fundamental signal. Therefore, assuming that the harmonics are stable, the effects caused by harmonic components that are integer multiples of the fundamental frequency tend to remain consistent within each segment of the time domain data 130.

[0051] However, unlike the first and second harmonics, the heart rate component is located at a position in the (original) spectrum 200-1 such that its corresponding data in the multiple segments of the time domain data 130 do not remain aligned between these segments. Depending on its position and intensity, this heart rate component may be completely averaged so as to no longer be visible in the average spectrum 200-2, or may be mostly averaged so as to be only partially visible in the average spectrum 200-2. In either case, the difference spectrum may include the maximum amplitude corresponding to the heart rate estimate.

[0052] Although the aforementioned techniques including determination of the difference spectrum and estimation of the heart rate based on its maximum amplitude may provide sufficiently accurate heart rate estimation, such estimation may be improved by further processing techniques on the time domain data 130. Figures 3 to 8 Discuss such techniques in more depth.

[0053] Figure 3A flow chart of exemplary pre-processing steps that may be applied prior to determining the spectrum 200-1 is provided. The pre-processing steps include five possible stages, or more specifically, five possible checks, which may be used in different combinations according to various embodiments. In any case where a check such as satisfying a required threshold or range is not achieved, the processing circuit device 110 may be configured to continue recording (measuring) other frames of the time domain data 130 to wait for the next frame as depicted in flowchart step 350. If each of the applicable checks is achieved, the processing circuit device 110 may be configured to continue further processing beyond pre-processing (i.e., perform core processing), as depicted in flowchart step 360. The first check may be a frame number check 310-1. For example, the processing circuit device 110 may automatically accept a certain number of frames of the time domain data 130 up to a pre-specified frame threshold 310-2. The frame threshold 310-2 may be 500 frames or another pre-specified number.

[0054] The pre-specified number of frames may correspond to a pre-specified amount of time (i.e., a time window) designated for receiving a received signal. More specifically, the time domain data 130 may correspond to a time window. For the example of an FMCW radar sensor, if the time window is 10 seconds, the time domain data 130 may correspond to 10 seconds of the radar system transmitting a frequency modulated continuous wave signal and receiving a corresponding echo (i.e., reflection) signal. For example, in this case, the device 100 may receive 500 frames in 10 seconds, or 1 frame every 0.02 seconds. A time window of 10 seconds may be selected for a desired balance. On the one hand, obtaining more time domain data may be beneficial for a more accurate average. On the other hand, the time window should not extend too long, in which case the random body movement of the target or other possible interference has a higher probability of making the time domain data 130 invalid. For example, in the case where the target is an infant, after 10 seconds, it becomes increasingly impossible for the infant to remain sufficiently still to obtain accurate measurements of valid time domain data 130. The time window may be adjusted according to each use case or user input, such as a range between 8 and 12 seconds.

[0055] The second check may be a noise check 320-1. The noise check 320-1 may be used to determine whether the time domain data 130 based on the received signal is below a pre-specified noise level threshold 320-2. For example, the processing circuit device 110 may be configured to apply a high pass filter for processing of the time domain data 130. Based on the high pass filter processing, the processing circuit device 110 may be configured to calculate the average amplitude of the high pass filtered time domain data 130 to obtain a noise level average. It may then be checked whether the noise level average is below the noise level threshold 320-2. Moving forward, the processing circuit device 110 may be configured to determine the spectrum 200-1 if the noise level average is below the pre-specified noise level threshold 320-2.

[0056] The third check may be a range check 330-1. The range check 330-1 may be used to determine whether the received signal includes sufficient time domain data from a pre-specified valid range 330-2. If it is determined that the time domain data 130 or the required amount of the time domain data 130 does not correspond to a pre-specified range, the processing circuit device 110 may be configured to record a new measurement of the new time domain data 130 (i.e., wait for the next frame 350 to make a new measurement). The range of the time domain data 130 may be determined by a distance discrete Fourier transform (DFT) of the time domain data 130. The distance-DFT may cause the device 100 to obtain a distance buffer divided into a plurality of distance bins. More specifically, the results of the distance-DFT may be stored in a memory of the device 100, wherein portions of the time domain data 130 are stored according to corresponding distance bins. The first distance bin may correspond to a first distance from the radar sensor, the second distance bin may correspond to a second distance from the radar sensor, and so on.

[0057] As part of or in addition to the range check 330 - 1 , a range-DFT may be performed on multiple frames of the time domain data 130 , which will refer to Figure 4 In addition, the two breathing checks 340-1a; 340-1b (i.e., breathing pattern checks) based on the corresponding breathing check verification 340-2a; 340-2b will refer to Figure 5 Provide explanation.

[0058] Figure 4A visual representation of exemplary processing steps for measuring the heart rate of a target in the field of view of a radar sensor, including the above-described range-DFT, is provided. As depicted, the processing circuit device 110 can obtain multiple frames 410-1; 410-2; 410-n, which are maintained as separate units of time domain data 130. The three points between the second frame and the nth frame are intended to express an arbitrary number of frames that can be part of the time domain data 130 between the second frame and the nth frame. As previously described, the number of frames can be 500 or another pre-specified number.

[0059] For example, the processing circuit device 110 may be configured to perform a mean removal of the time domain data 130. More specifically, the processing circuit device 110 may be configured to calculate an average of the amplitude values ​​of the time domain data 130 and subtract the average from the amplitude values ​​to generate the mean removed time domain data. This may shift the previous values ​​(which may include an offset) to be centered about zero. In general, using a form of mean removed time domain data 130 may enable more accurate generation of the spectra 200-1; 200-2. If applicable, the mean removed version of the time domain data 130 may also be used to more accurately perform further checks, such as confirming the target's effective breathing rate and heart rate (in Figure 5 (available in ).

[0060] Mean removal can be applied to the entire time domain data 130, or on a per-frame basis. Using a per-frame basis, the average of the amplitude values ​​can be the average of a given frame, which can be subtracted from the amplitude values ​​of the given frame. This can generate mean-removed time domain data specifically for a given frame. As depicted, a first portion 412-1 of mean-removed time domain data can correspond to a first time frame 410-1, a second portion 412-2 of mean-removed time domain data can correspond to a second time frame 410-2, and so on until the nth portion 412-n for the nth frame 410-n. In addition, the distance-DFT described previously can be performed using the respective portions of mean-removed time domain data. Further reference to the processing of the time domain data 130 can be applied to mean-removed or non-mean-removed time domain data 130.

[0061] The processing circuitry 110 may be configured to perform a separate distance-DFT on each frame 410-1; 410-2; 410-n of the time domain data 130 (i.e., on a per-frame basis). If mean removal of the time domain data is applied, a distance-DFT may be performed on each portion of the mean-removed time domain data 412-1; 412-2; 412-n corresponding to its respective frame. Thus, the distance-DFT may be referred to as a short-time distance-DFT, which is applied to the short-time distance data (respective frames or portions) of the time domain data 130. Each of the respective distance-DFTs performed on the respective frames or portions of the (mean-removed) time domain data 130 may result in the generation of a respective distance buffer 420-1; 420-2; 420-n, as depicted, which corresponds to Figure 4 In the case where the number of frames for a complete iteration of the measurement is pre-specified as 500 frames (i.e., n=500), the processing circuit device 110 may perform 500 corresponding distance-DFTs on 500 frames or portions of the (averaged) time domain data 130 to obtain 500 corresponding buffers. Each of the distance buffers may include an equal number of distance bins, such that a particular distance bin of a first distance buffer may have a corresponding distance bin at an equivalent position or with an equivalent index within a second distance buffer.

[0062] The processing circuit device 110 may utilize the maximum value (e.g., the maximum amplitude value or the average amplitude value) of the (average removed) time domain data 130 to identify a specific distance bin for each distance buffer. More specifically, a set of corresponding distance bins spanning multiple (e.g., 500) distance buffers may be selected, wherein the set of distance bins each corresponds to an equivalent index or position within its corresponding distance buffer. The processing circuit device 110 may be configured to select such a set of distance bins corresponding to a common position or index, each (or most) of the distance bins in the distance bins including the maximum amplitude of the (average removed) time domain data 130. Therefore, the set of corresponding (i.e., similarly positioned) distance bins may also be simply referred to as the "range bin" or "identified distance bin" of the set, specifically referring to the common position within each distance buffer. Figure 4 The identified range bins 422 corresponding to the maximum amplitude of the (average removed) time domain data 130 are shown.

[0063] The processing circuit device 110 can be configured to check whether the identified range bin 422 is within the selection of pre-specified range bins for the corresponding range buffer. The selection of the pre-specified range bins can correspond to a measurement range within an acceptable distance range from the radar sensor. Such an acceptable distance range can be, for example, between 1 and 1.5 meters, or another acceptable range of similar scale (e.g., between 0.5 meters and 3 meters). If the identified range bin 422 is not within the selection of the pre-specified range bins, the time domain data 130 can be considered invalid, and the radar system can start a new iteration of measurement (i.e., wait for the next frame 350). On the other hand, if the identified range bin 422 is indeed within the selection of the pre-specified range bins, the time domain data 130 can be considered to be within the pre-specified valid range 330-2.

[0064] Based on the determination of the valid range of the time domain data 130, for each distance buffer 420-1; 420-32; 420-n, the processing circuit device 110 can determine the spectrum 200-1 and the average spectrum 200-2 for the (removed average) time domain data 130 corresponding to the identified distance bin 422. In other words, the processing circuit device 110 can be configured to transform the (removed average) time domain data 130 by distance-DFT, select a portion of the generated data, and determine the spectrum 200-1 based on the selected data, which corresponds to a certain portion of the time domain data 130.

[0065] Since the range-DFT can be applied across a range buffer, it can be considered to be applied to long time domain data.By focusing the Doppler-DFT to the identified range bins 422, the spectrum 200-1; 200-2 can be determined more accurately.

[0066] Figure 4 The processing steps presented in, including the range-DFT applied to (removed average) time domain data, can generate a (removed average) distance-time dataset 430. The distance-time dataset 430 is depicted as including data within the identified range bins 422. Based on the identified range bins 422, the distance-time dataset 430 can be extracted for further inspection 340-1a; 340-1b.

[0067] Figure 5An exemplary distance-time curve 500 corresponding to the distance-time data set 430 is depicted. The distance-time curve 500 can be used to check whether there is a consistent pattern corresponding to the breathing of the target (i.e., a breathing pattern). Such a breathing pattern check 340-1a; 340-1b can be performed before the Doppler-DFT, wherein the Doppler-DFT is performed only if the breathing pattern check 340-1a; 340-1b passes the corresponding threshold 340-2a; 340-2b (see Figure 3 ).

[0068] Breathing pattern checks 340-1a; 340-1b, which may be applied to the time domain data 130, or specifically, to the de-averaged time domain data corresponding to the identified range bins 422 as previously described. Alternatively, the processing circuit device 110 may be configured to perform de-averaging after the range-DFT, by calculating the average of the amplitude values ​​in the identified range bins 422 and subtracting the average from the average of the amplitude values ​​in the identified range bins 422. This may shift the information of the time domain data 130 corresponding to the identified range bins 422 from previous values ​​to values ​​centered around zero to generate a de-averaged range-time data set 430 for the identified range bins 422. The de-averaged range-time data set 430 may enable faster processing of further checks of the data to ensure that valid breathing rates and heart rates can be found for targets in the field of view of the radar sensor. If such checks are passed, the spectrum may be specifically determined based on the de-averaged range-time data set 430.

[0069] An exemplary distance-time curve 500 based on removing the average distance-time data set 430 (where the average removal is performed before or after the distance-DFT) intuitively shows how this shifts the time domain data 130 to be centered around zero. The x-axis of the distance-time curve 500 can correspond to any arbitrary time unit (e.g., seconds), while the y-axis of the distance-time curve 500 can correspond to a complex amplitude proportional to the range (i.e., distance or displacement), which can be measured using its appropriate units.

[0070] Indeed, when plotted, the range-time curve 500 may provide a visual depiction of a breathing pattern. The target's breathing may include inhalation and exhalation, which may repeat in a stable pattern over time. For example, when inhaling, air is taken in and the target's chest may expand, which may thereby decrease the range (i.e., distance or displacement) between the target and the radar sensor. On the other hand, when exhaling, air is exhaled and the target's chest may contract, which may thereby increase the range (i.e., distance or displacement) between the target and the radar sensor. Since the breathing pattern must follow a recognizable pattern, certain checks may be implemented to ensure that the target is breathing correctly. In the case of consistent breathing, the range may be reliably performed. Figure 1 The heart rate is estimated by following the processing steps outlined in FIG2 .

[0071] The removed average distance-time data set 430 may correspond to a removed average distance-time curve 500, which may depict a change in range (distance or displacement) that may be examined for consistency. Figure 3 In the first breath check 340 - 1 a presented in , the processing circuit device 110 may be configured to check whether the peak-to-peak value for the removed average distance-time dataset 430 is within pre-specified limits based on the average of multiple previous peak-to-peak values ​​in the removed average distance-time dataset 430 .

[0072] For example, the first peak-to-peak value corresponding to the increase in range (distance or displacement) starting at time t=0 may be approximately 4.0×10 -3 The second peak-to-peak value corresponding to the decrease in range (distance or displacement) may be approximately 3.5×10 -3 The third peak-to-peak value, the fourth peak-to-peak value, and the fifth peak-to-peak value may also be approximately 3.5×10 -3 units, and for the third decrease in range (distance or displacement), the sixth peak-to-peak value may be approximately 4×10 -3 units. Multiple peak-to-peak values ​​may be checked. Alternatively, each peak-to-peak value may be checked, wherein the peak-to-peak value is determined to correspond to any increase from a local minimum to a local maximum, or to any decrease from a local maximum to a local minimum (where local minima or local maxima of very narrow frequency ranges may be ignored). For all peak-to-peak values, the peak-to-peak threshold 340-2a (at Figure 3340-2a) may be determined based on a plurality or all of the peak-to-peak values, such as by taking an average. Processing circuitry 110 may be configured to determine that the spectrum of averaged distance-time dataset 430 is removed if all or a majority of the peak-to-peak values ​​(as described above) are within the limits of peak-to-peak threshold 340-2a.

[0073] exist Figure 3 In the second breath check 340-1b in, the processing circuit device 110 can be configured to determine a plurality of derivative values ​​corresponding to the rate of change of the distance at the corresponding time of the average distance-time data set 430 (or the distance-time curve 500). Each derivative can be determined based on a time instance, which can be within a time interval or correspond to a time interval. The x-axis of the distance-time curve 500 can correspond to the time in seconds. Therefore, the distance-time curve 500 can correspond to the previously mentioned 10 second time window. Therefore, each complete cycle from a local minimum (or maximum) to an adjacent local minimum (or maximum) can correspond to a time of 1.33 seconds for a complete breath, which is a time interval corresponding to a breathing rate frequency of 0.75 Hz or a breathing rate of 45 times per minute. Within the 10 second time window, 8 inhalations (increases in distance / displacement) and 7 exhalations are depicted, corresponding to approximately 7.5 breaths (0.75 Hz) in 10 seconds.

[0074] Each inhalation may have a derivative value associated with it. For example, the derivative value may correspond to the change in range (distance or displacement) of the data set from a local minimum to a local maximum divided by the change in time in seconds. Similarly, the same is true for the change from a local maximum to a local minimum (where local minima or local maxima of very narrow frequency ranges may be ignored, as described above). Alternatively or additionally, the derivative value may correspond to the change in distance (displacement) divided by the corresponding change in time, where the time interval within the distance-time curve 500 for the corresponding distance and time changes may be arbitrarily selected. The processing circuit device 110 may be configured to determine any number of derivative values ​​within the distance-time curve 500, which may be from a local minimum to a local maximum or vice versa, or the derivative value may correspond to a narrower time interval within these points.

[0075] The processing circuitry 110 may be configured to determine the variance of such derivative values ​​as the derivative threshold 340-2b (at Figure 3340 - 2 b), which may be based on the absolute values ​​of the determined derivatives. In addition, processing circuit device 110 may check that all or most of the determined derivatives (or their absolute values) are within derivative threshold 340 - 2 b. If each derivative value is within derivative threshold 340 - 2 b, processing circuit device 110 may determine spectrum 200 - 1 for removing average distance-time data set 430.

[0076] In general, the above five checks 310-1; 320-1; 330-1; 340-1a; 340-1b may be used in any combination to ensure, with the aid of their respective thresholds 310-2; 320-2; 330-2; 340-2a; 340-2b, that the time domain data 130 is valid data, which may be considered reliable for determining the breathing rate and heart rate of the target within the field of view of the radar sensor. Such checks may be part of the pre-processing of the device 100. If all applied checks are passed, the processing of the device 100 may continue to its core processing steps, such as Figure 1 and as outlined in Figure 2. In addition to the core processing steps outlined previously, the core processing may include further processing features which will be referred to in Figure 6 Provide a more detailed presentation.

[0077] Figure 6 A flow chart is provided of exemplary optional processing steps 610; 620; 630; 640 that may be used to measure the heart rate of a target in a field of view of a radar sensor.

[0078] Figure 6 The first optional processing step shown is to use a finite impulse response filter to denoise 610. The processing circuit device 110 can be configured to determine a spectrum for the time domain data, and then apply a finite impulse response filter to the amplitude values ​​of the spectrum. Finite impulse response FIR filters are generally digital filters used in signal processing. FIR filters can be used, particularly as a specific form of high pass filter, to eliminate signals corresponding to vibrating objects in the background of targets that may be present in the field of view of the radar sensor. The vibrations may include switching back and forth between a positive direction and a negative direction of movement through space. In other words, the vibrations may switch back and forth between having a positive velocity in a positive direction and having a negative velocity in a negative direction, which may be recorded as part of the time domain data 130 and may interfere with the signal processing of the time domain data 130. For example, air conditioners are known to be vibrating objects that may interfere with the signal processing techniques of radar systems, as well as other possible causes.

[0079] Although FIR filters may be conventionally or generally used to filter time domain data prior to Doppler-DFT (i.e., filtering by data processing), processing circuit device 110 may be specifically configured to apply FIR filters to the data of generated spectrum 200-1 (i.e., the time domain data 130 in a processed form as part of spectrum 200-1) after Doppler-DFT to prevent interference of vibrating objects with signal processing. For example, for a wave signal transmitted by a frequency modulated continuous wave FMCW radar, the frequency components in the Doppler-DFT analysis may correspond to the speed or velocity of an object in the field of view of the radar sensor, which is derived from the Doppler effect, describing how the frequency of the transmitted and reflected waves may change when reflected from an object having relative motion with the radar sensor.

[0080] Although an FIR filter (such as an FIR high-pass filter) can filter data in the time domain before Doppler-DFT based on high speed or specific speed components within the signal, an FIR filter applied to filter data in the frequency domain after Doppler-DFT can filter high speed changes or specific speed change components within the data. Thus, interference in the data caused by vibration can be filtered out, which can correspond to specific speed changes that cause movement in the positive and negative directions previously described. In general, various versions of FIR filters can be applied to filter the effects of various vibrating objects in the field of view of a radar sensor, particularly for the vibration range of common household objects such as air conditioning units.

[0081] Figure 6 The second optional processing step shown involves checking the validity 620 of the spectrum 200-1 and the average spectrum 200-2. In the first check, the processing circuit device 110 can be configured to determine the breathing rate frequency by detecting a first frequency, which corresponds to the maximum amplitude value of the spectrum 200-1, and detecting a second frequency, which corresponds to a local maximum of amplitude values ​​in the spectrum 200-1 within a pre-specified frequency range starting from twice the first frequency (e.g., the first harmonic). Now returning to reference Figure 2 and the depicted (original) spectrum 200-1, the breathing rate frequency can be detected at its maximum amplitude 210-1 corresponding to the breathing rate (e.g., 45 breaths per minute), and subsequent local maxima can be detected within a pre-specified range of twice the breathing rate frequency (e.g., 90 breaths per minute). In this particular case, the pre-specified frequency range for subsequent local maxima can be between 87 and 93 breaths per minute, or another frequency range based on accuracy or precision requirements, which frequency range can be adjustable.

[0082] In addition, the processing circuit device 110 can be configured to detect a third frequency corresponding to another local maximum of the amplitude value in the spectrum, specifically, within a second pre-specified frequency range starting from three times the first frequency (i.e., the second harmonic). In the (original) spectrum 200-1 of Figure 2, a second subsequent local maximum from the respiratory rate frequency can be detected at 135 breaths per minute or within a second pre-specified range therefrom. In this particular case, the second pre-specified range can be 132 to 138 breaths per minute, or it can also be another frequency range adjusted based on accuracy or precision requirements. In particular, for the second harmonic of the respiratory rate frequency, the pre-specified range can be narrow enough to distinguish the second harmonic from the local maximum corresponding to the heart rate frequency. If the frequency value of the first harmonic or the second harmonic is not within the corresponding pre-specified range, the processing circuit device 110 can be configured to mark the time domain data 130 as invalid, cancel further processing of the time domain data 130, and start a new iteration process for recording time domain data.

[0083] Such validity tests may also be performed with respect to the average spectrum 200-2 in FIG. 2 . For example, the processing circuit device 110 may be configured to verify that the maximum amplitude value of the average spectrum 200-2 corresponds to a frequency within a pre-specified frequency range from the respiratory rate frequency. Such a check may provide an accurate assessment of whether the respiratory rate frequency determined based on the maximum amplitude of the (original) spectrum 200-1 is accurate. In the case of an inaccurate respiratory rate frequency, the time domain data 130 may not be divided into multiple segments with sufficiently accurate time intervals, which are the inverse of the somewhat inaccurate respiratory rate frequency. This may result in the maximum amplitude frequency being shifted from the respiratory rate frequency in the (original) spectrum 200-1 to the frequency in the average spectrum 200-2. For example, such a pre-specified range may be between 42 and 48 breaths per minute, or it may be adjusted to another range according to accuracy and precision requirements. If the maximum amplitude value of the average spectrum 200-2 corresponds to a frequency within its pre-specified frequency range from the respiratory rate frequency, later processing steps for determining a differential spectrum and estimating the heart rate may be performed.

[0084] If this is not the case, the processing circuit device 110 can be configured to determine a new breathing rate frequency based on adjacent local maxima of the amplitude values ​​in the (original) spectrum 200-1. For example, there may be highly localized spikes in the determined (original) spectrum 200-1, which may have caused the value assigned to the breathing rate frequency to be somewhat inaccurate. To prevent this, the processing circuit device 100 can be configured to accept breathing rate frequencies corresponding to amplitudes within a certain pre-specified range. If the breathing rate frequency is determined to be inaccurate (for any reason), the processing circuit device 110 can be configured to repeat its previously performed calculations for subsequent or previous local maxima, starting from the previously marked breathing rate frequency (wherein local minima or local maxima of very narrow frequency ranges may be ignored, as described above).

[0085] Such local maxima may also be searched within a customized frequency range that may correspond to a breathing rate corresponding to a physical characteristic of the target. For example, a target that is a human newborn infant may have a known breathing rate frequency range, given certain assumptions (e.g., being at rest). A newborn infant may have a search range that varies between 20 and 70 breaths per minute, a larger infant / toddler may have a search range that varies between 15 and 40 breaths per minute, and an adult may have a search range that varies between 8 and 25 breaths per minute. Other physical characteristics besides the target age may also affect the customized range. Such a configuration may be adjustable. The adjustable configuration may be adjusted by user input received through a user interface and other possible means.

[0086] The processing circuit device 110 can be configured to repeat the same processing steps for the new breathing rate frequency, such as for subsequent or previous local maxima, to obtain a new average spectrum based on the new breathing rate frequency. In addition, the previously described check of the maximum amplitude value of the new average spectrum can be verified to correspond to its frequency within a pre-specified frequency range starting from the new breathing rate frequency. If this is not the case, it can be checked whether another local maximum of the (original) spectrum 200-1 is within a customized frequency range based on the physical characteristics of the target, and the previously described steps are repeated. If no local maximum within the customized frequency range satisfies the required checks, the processing circuit device 110 can be configured to mark the time domain data 130 as invalid, cancel further processing thereof, and start a new iteration of the process for recording the time domain data.

[0087] Although a number of examples related to calculating thresholds have been explained in depth, the processing circuit device 110 may be configured to apply various thresholds for ensuring that accurate values ​​of the respiratory rate frequency and the heart rate frequency are determined / estimated. Such thresholds may be applied in various combinations and at various stages of processing. Figure 6 As depicted, a further example of processing the time domain data 130 to improve the accuracy and / or precision of the heart rate estimate and possibly the respiration rate estimate is the application of a correlation function 630 (which will be referred to in Fig. 7A , Figure 7B ,and Figure 7C ) and using a recursive model 640 for heart rate estimation (will refer to Figure 8 for discussion).

[0088] Fig. 7A An exemplary (original) spectrum 200-1 (with a heart rate component) and an exemplary average spectrum 200-2 (without a heart rate component) are depicted plotted together. They are each provided in the context of an exemplary non-correlated spectrum 730-1. If an accurate value for the respiratory rate frequency is selected, the maximum amplitudes of the two spectra, as well as the local maxima of the corresponding first and second harmonics, are highly overlapped, as shown in the non-correlated spectrum 730-1. However, in order to be able to further increase the accuracy and / or precision of the heart rate estimate, both the (original) spectrum 200-1 and the average spectrum 200-2 can be subjected to a correlation operation.

[0089] More specifically, the processing circuit device 110 may be configured to perform a first correlation operation on the (original) spectrum 200-1 to obtain a correlated (original) spectrum (in Figure 7C as 730-3a), and for the average spectrum 200-2 (in Figure 7C 730-3b) performs a second correlation operation to obtain a correlated average spectrum (in Figure 7C Each correlation operation may be between an original version and a windowed version of the corresponding spectrum, wherein the corresponding windowed version is changed by applying a first Gaussian window centered at a first frequency corresponding to a maximum amplitude of the spectrum, applying a second Gaussian window centered at a second frequency twice the first frequency, applying a third Gaussian window centered at a third frequency three times the first frequency, and invalidating all amplitude values ​​outside the three Gaussian windows.

[0090] For example, in Fig. 7AIn, (original) spectrum 200-1 is depicted as having its first Gaussian window, second Gaussian window, and third Gaussian window. The first Gaussian window for (original) spectrum 200-1 is determined to correspond to the vertical lines 732-1a and 732-1b depicted. Each vertical line is located at the frequency corresponding to the local minimum immediately before and immediately after the maximum amplitude, respectively. The second Gaussian window for (original) spectrum 200-1 is provided by the vertical lines 732-2a and 732-2b depicted, and is also provided by the local minimum immediately before and immediately after the second highest local maximum in (original) spectrum 200-1. The third Gaussian window for (original) spectrum 200-1 is provided by the vertical lines 732-3a and 732-3b depicted, and is determined in a similar manner.

[0091] The first Gaussian window, the second Gaussian window, and / or the third Gaussian window may also be determined in different ways, such as by a pre-specified frequency range from the corresponding maximum value, which may be customized. Although the corresponding first Gaussian window, the second Gaussian window, and the third Gaussian window of the average spectrum 200-2 may also be used for its related operations, for the sake of clarity, in Fig. 7A They are not depicted in the .

[0092] Once the first Gaussian window, the second Gaussian window, and the third Gaussian window are determined, they can be applied to generate the windowed (original) spectrum 730-2, as shown in FIG. Figure 7B Depicted. The first correlation operation can be between the (original) spectrum 200-1 and the windowed (original) spectrum 730-2. Generally speaking, in the correlation operation, two data points from the two related spectrums can be multiplied and added to the recursive sum in the first correlation iteration of the correlation operation. This process can be repeated for the second iteration of the correlation operation, wherein one of the spectra is shifted relative to the other spectrum to determine a new set of data points to be correlated, and the new product is determined and added to the recursive sum. Such iterations can be repeated according to further equivalent shifts to determine the first correlation spectrum 730-3a, which can depict the result of the first correlation operation performed between the (original) spectrum 200-1 and the windowed (original) spectrum 730-2.

[0093] In addition, for the second correlation operation described previously, the above technique can be repeated. The second correlation operation can be between the average spectrum 200-2 and the windowed version of the average spectrum (not depicted, similar to 730-2), and the windowed version of the average spectrum is similar to the windowed version of (original) spectrum 200-1. In other words, the average spectrum can also apply the first Gaussian window, the second Gaussian window, and the third Gaussian window (not depicted, similar to 732-1a; 732-1b; 732-2a; 732-2b; 732-3a; 732-3b), as described above, to form its windowed version. Then, also as described above, the correlation operation can be performed to obtain the corresponding correlation spectrum 730-3b.

[0094] Figure 7C An exemplary correlation spectrum 730-3 is depicted, including an exemplary correlation spectrum 730-3a and an exemplary correlation average spectrum 730-3b after respective correlation operations. Once the two correlation spectra 730-3a; 730-3b are determined, they may include greater similarity within the windowed portion, while portions of the spectra 730-3a; 730-3b outside the windowed portion may be suppressed to lower amplitude values, such as by Fig. 7A The non-correlation spectrum of 730-1 and Figure 7C The result of correlation spectrum 730-3 shows that the correlation (original) spectrum 730-3a and the correlation average spectrum 730-3b overlap more accurately, which enables the corresponding respiration rate component to be more effectively eliminated in the difference spectrum. Therefore, it is possible to more accurately determine Figure 2A The heart rate component 212 in FIG.

[0095] Figure 8 Provides Figure 7C The same exemplary correlation spectrum 730-3 as in Figure 8 The difference spectrum 200-3 is determined by subtracting the average spectrum 200-2 from the (original) spectrum 200-1. Specifically, the processing circuit device 110 can be configured to subtract the relevant average spectrum 730-3b from the relevant (original) spectrum 730-3a to obtain the difference spectrum 200-3, where the subtraction of one spectrum from the other follows: Figure 1 The previous definition provided. Since the corresponding correlation spectrum emphasizes the frequency components corresponding to the breathing rate frequency, as well as its first harmonic components and second harmonic components, it can enable the difference spectrum 200-3 to be determined more accurately than if the correlation spectrum is not used. In addition, the heart rate can be estimated to correspond to the maximum amplitude 230-1 of the difference spectrum 200-3, as depicted. The difference spectrum 200-3 is depicted using an enlarged scale and shows the heart rate components at the frequency of the heart rate (shown with greater accuracy as approximately 128 beats per minute). This is shown by Figure 2A The heart rate component 212 is depicted in FIG.

[0096] In addition, the processing circuit device 110 can be configured to apply a window to the difference spectrum 200-3 and search for the maximum amplitude 230-1 within the window of the difference spectrum 200-3 for estimating the heart rate frequency. The window of the difference spectrum 200-3 can specify a frequency range corresponding to a range of possible heart rates of the target, which can also be customized according to the targets in the field of view of the radar sensor (e.g., according to user input).

[0097] The processing circuit device 110 may also be configured to record multiple heart rate estimates within a single recording iteration or within briefly separated recording iterations. For example, given a time window of 10 seconds or a similar value (e.g., 8 to 12 seconds) for measurement iterations, approximately 5 to 7 estimates of the target's heart rate may be performed over a one-minute period. If the target is sufficiently stationary and has minimal random body movement, all estimates for the corresponding measurement iteration may be determined to be valid. Alternatively, if too much random body movement occurs within the time window, thereby preventing the measurement of valid data for the heart rate estimate, the time domain data corresponding to that particular time window may be discarded and further measurement iterations may be continued. For example, the processing circuit device 110 may be configured to stop measuring and immediately start a new measurement iteration (e.g., of 10 seconds) if a threshold associated with random body movement has been exceeded, which may include Figure 3 The checks and corresponding thresholds described in .

[0098] In particular, the processing circuit device 110 can be configured to perform multiple heart rate estimates, where each estimate is applied to an iteration of the recursive model to obtain an accurate heart rate estimate. For example, a first estimate of the heart rate of a target can be based on a first heart rate estimate with a first error range. Given a second heart rate estimate of the same target immediately following the first heart rate estimate, where both estimates are based on corresponding time domain data 130 under the same circumstances, the second heart rate can increase the confidence of the previous estimate. In addition, the recursive model can determine that the heart rate of the target is a value within a smaller error range.

[0099] The recursive model can continue to be applied, where further estimates of the target's heart rate can be given, such as five or more estimates. Based on each successive estimate, the recursive model can determine that the probability of the target's heart rate being within a narrowing range increases. In general, such a narrowing range can be part of a refined heart rate estimate, where the final estimate is considered to be more accurate and / or more precise.

[0100] In addition to accurate heart rate estimation, the device 100 can also be used to increase the confidence of the respiratory rate estimate (i.e., respiratory rate determination). Based on many possible checks that apply the respiratory rate frequency and its first and second harmonics, a higher confidence in the respiratory rate can be obtained. In addition, other possible checks (e.g., checks related to heart rate) can also be completed and verified, thereby further increasing the confidence in the respiratory rate, which other possible checks are not even possible without an accurate value for the respiratory rate.

[0101] In general, the device 100 used according to a configuration in the above-mentioned various configurations can provide the specific benefit of long-term monitoring of the biological signs of the same target. In the specific case of a newborn, heart rate and respiratory rate can be measured and recorded for many times, and performed in many different days (or longer time spans), which can provide nursing staff with more reliable information related to the biological activity and overall health of the newborn. Such benefits are also possible for larger children and adults. Further benefits can be a better understanding of the pattern of this biological activity in various target types (for example, human infants or adults, or also animals), and how they may be affected, particularly if other events occurring in the associated time span are also recorded. In general, device 100 provides a new way for checking biological activity and health by using the new technology of radar system.

[0102] Fig. 9 An example of a radar system 900 is illustrated, which includes an apparatus 910 as described herein, such as apparatus 100, and a radar sensor 920 configured to generate a received signal. Radar sensor 920 may be, for example, a FMCW (Frequency Modulated Continuous Wave) radar sensor.

[0103] Although device 910 and radar sensor 920 are Fig. 9 910 is depicted as a separate block, but in other examples, the device 910 may be partially or fully included in the radar sensor 920. Therefore, the radar sensor 920 may include the device 910. In the case where the device 910 is only partially included in the radar sensor 920, the radar system 900 may perform distributed processing that implements corresponding parts of the processing steps. In the case where the device 910 is integrated in the radar sensor 920, the control circuit device and the radar sensor 920 may be jointly integrated (embedded) in a single semiconductor chip, or integrated (embedded) in more than one semiconductor chip.

[0104] Further details and aspects of the radar system 900 will be explained in conjunction with the above-mentioned techniques or one or more examples, such as reference Figure 1The radar system 900 may include one or more additional optional features corresponding to one or more aspects of the proposed technology, or one or more examples described above.

[0105] Fig.10 A flow chart of an example of a (e.g., computer-implemented) method 1000 is illustrated. The method 1000 may be performed by an apparatus described herein, such as the apparatus 100. The method 1000 includes determining 1010 a frequency spectrum corresponding to the time domain data based on a received signal. The method 1000 also includes determining 1020 a respiratory rate frequency based on a maximum amplitude of the frequency spectrum. In addition, the method 1000 includes determining 1030 a plurality of segment frequency spectra based on the respiratory rate frequency, each segment frequency spectrum corresponding to a respective segment of the time domain data, and determining 1040 a mean of the segment frequency spectra to obtain an average frequency spectrum. The method also includes subtracting 1050 the average frequency spectrum from the frequency spectrum to obtain a differential frequency spectrum, and estimating 1060 a heart rate of the target based on the maximum amplitude of the differential frequency spectrum.

[0106] More details and aspects of method 1000 will be explained in conjunction with the above-mentioned techniques or one or more examples, such as reference Figure 1 The method 1000 may include one or more additional optional features corresponding to one or more aspects of the proposed technology, or one or more examples described above.

[0107] In the following, some examples of the proposed technique are presented:

[0108] One example (e.g., Example 1) relates to a device comprising a processing circuit device configured to determine a spectrum corresponding to time domain data based on a received signal; determine a respiratory rate frequency based on a maximum amplitude of the spectrum; determine a plurality of segment spectra based on the respiratory rate frequency, each segment spectrum corresponding to a corresponding segment of the time domain data; determine a mean of the segment spectra to obtain an average spectrum; subtract the average spectrum from the spectrum to obtain a differential spectrum; and estimate the heart rate of the target based on the maximum amplitude of the differential spectrum.

[0109] Another example (e.g., Example 2) relates to the previous example (e.g., Example 1) or any other example, and also includes: the processing circuit device is configured to apply a high-pass filter for processing time domain data; calculate the average amplitude of the high-pass filtered time domain data to obtain a noise level average; and if the noise level average is lower than a pre-specified noise level threshold, determine the frequency spectrum.

[0110] Another example (e.g., Example 3) relates to a previous example (e.g., one of Examples 1 or 2) or any other example, and also includes: the processing circuit device is configured to calculate an average of the amplitude values ​​of the time domain data; subtract the average from the amplitude values ​​of the time domain data to generate time domain data without the average; and determine a spectrum for the time domain data without the average.

[0111] Another example (e.g., Example 4) relates to a previous example (e.g., one of Examples 1 to 3) or any other example, and also includes: the processing circuit device is configured to perform one or more distance Fourier transforms of time domain data to obtain one or more distance buffers, each distance buffer being divided into multiple distance bins; for each distance buffer, identifying the distance bin associated with the maximum amplitude to determine a distance-time data set; and determining a frequency spectrum for the distance-time data set corresponding to the identified distance bin.

[0112] Another example (e.g., Example 5) relates to the previous example (e.g., Example 4) or any other example, and also includes: the processing circuit device is configured to check whether the peak-to-peak value for the distance-time dataset is within pre-specified limits based on the average of multiple previous peak-to-peak values ​​in the distance-time dataset; and if the peak-to-peak value is within its pre-specified limits, determine the frequency spectrum for the distance-time dataset.

[0113] Another example (e.g., Example 6) relates to a previous example (e.g., one of Examples 4 or 5) or any other example, and also includes: the processing circuit device is configured to determine a plurality of derivative values ​​corresponding to the rate of change of the distance at corresponding times of the distance-time dataset; determine the variance of the plurality of derivative values; check whether each of the derivative values ​​is within a threshold determined by the variance; and if each derivative value is within the threshold, determine a frequency spectrum for the distance-time dataset.

[0114] Another example (e.g., Example 7) relates to a previous example (e.g., one of Examples 1 to 6) or any other example, and also includes: the processing circuit device is configured to determine a spectrum for the time domain data and then apply a finite impulse response filter to the amplitude values ​​of the spectrum.

[0115] Another example (e.g., Example 8) relates to a previous example (e.g., one of Examples 1 to 7) or any other example, and also includes: the processing circuit device is configured to determine the breathing rate frequency by detecting a first frequency and detecting a second frequency, wherein the first frequency corresponds to the maximum amplitude value of the spectrum, and the second frequency corresponds to the local maximum of the amplitude value in the spectrum within a pre-specified frequency range starting from twice the first frequency.

[0116] Another example (e.g., Example 9) relates to the previous example (e.g., Example 8) or any other example, and also includes: the processing circuit device is configured to verify that the maximum amplitude value of the average spectrum corresponds to a frequency within a pre-specified frequency range starting from the breathing rate frequency.

[0117] Another example (e.g., Example 10) relates to the previous example (e.g., Example 9) or any other example, and also includes: for a case where the maximum amplitude value of the average spectrum corresponds to a frequency that is not within a pre-specified frequency range starting from the breathing rate frequency, the processing circuit device is configured to determine a new breathing rate frequency based on adjacent local maxima of the amplitude values ​​in the spectrum; obtain a new average spectrum based on the new breathing rate frequency; and verify that the maximum amplitude value of the new average spectrum corresponds to a frequency within a pre-specified frequency range starting from the new breathing rate frequency.

[0118] Another example (e.g., Example 11) relates to a previous example (e.g., one of Examples 1 to 10) or any other example, and also includes: a processing circuit device is configured to perform a first correlation operation on the spectrum to determine a correlated spectrum, and to perform a second correlation operation on the average spectrum to determine a correlated average spectrum, wherein each correlation operation is between an original version and a windowed version of the corresponding spectrum, wherein the corresponding windowed version is changed by applying a first Gaussian window centered on a first frequency corresponding to the maximum amplitude of the spectrum, applying a second Gaussian window centered on a second frequency twice the first frequency, applying a third Gaussian window centered on a third frequency three times the first frequency, and invalidating all amplitude values ​​outside the three Gaussian windows; and the correlated average spectrum is subtracted from the correlated spectrum to obtain a difference spectrum.

[0119] Another example (e.g., Example 12) relates to a previous example (e.g., one of Examples 1 to 11) or any other example, and further includes: the processing circuit device is configured to perform multiple heart rate estimates, where each estimate is applied to an iteration of the recursive model to obtain an accurate heart rate estimate.

[0120] One example (e.g., Example 13) relates to a radar system comprising an apparatus according to a previous example (e.g., any one of Examples 1 to 12) or any other example, and a radar sensor configured to generate a received signal.

[0121] An example (e.g., Example 14) relates to a method comprising determining a spectrum corresponding to time domain data based on a received signal; determining a respiratory rate frequency based on a maximum amplitude of the spectrum; determining multiple segment spectra based on the respiratory rate frequency, each segment spectrum corresponding to a corresponding segment of the time domain data; determining a mean of the segment spectra to obtain an average spectrum; subtracting the average spectrum from the spectrum to obtain a difference spectrum; and estimating the heart rate of the target based on the maximum amplitude of the difference spectrum.

[0122] One example (eg, Example 15) relates to a program having a program code for performing a method according to a previous example (eg, Example 14) when the program is executed on a processor or programmable hardware.

[0123] The aspects and features described with respect to a specific one of the previous examples may also be combined with one or more other examples to replace the completely identical or similar features of the other examples, or to additionally introduce these features into the other examples.

[0124] Examples may also be or relate to (computer) programs including program codes, which, when executed on a computer, processor or other programmable hardware component, are used to perform one or more of the above methods. Therefore, the steps, operations or processes of the above different methods may also be performed by a programmed computer, processor or other programmable hardware component. Examples may also cover program storage devices such as digital data storage media, which are machine-readable, processor-readable or computer-readable, and encode and / or contain machine-executable, processor-executable or computer-executable programs and instructions. For example, program storage devices may include or be digital storage devices, magnetic storage media such as disks and tapes, hard drives, or optically readable digital data storage media. Other examples may also include computers, processors, control units, (field) programmable logic arrays ((F) PLAs), (field) programmable gate arrays ((F) PGAs), graphics processor units (GPUs), application specific integrated circuits (ASICs), integrated circuits (ICs) or system-on-chip (SoC) systems programmed to perform the steps of the above methods.

[0125] It should also be understood that the disclosure of several steps, processes, operations or functions disclosed in the specification or claims should not be interpreted as implying that these operations necessarily depend on the order described, unless explicitly stated in individual cases or necessary for technical reasons. Therefore, the previous description does not limit the execution of several steps or functions to a certain order. In addition, in other examples, a single step, function, process or operation may include and / or be decomposed into several sub-steps, sub-functions, sub-processes or sub-operations.

[0126] If some aspects have been described with respect to a device or system, these aspects should also be understood as descriptions of corresponding methods. For example, functional aspects of a block, device, or device or system may correspond to features of a corresponding method, such as method steps. Therefore, aspects described with respect to a method should also be understood as descriptions of characteristics or functional features of a corresponding block, corresponding element, corresponding device, or corresponding system.

[0127] The following claims are hereby incorporated into the detailed description, where each claim may stand on its own as a separate example. It should also be noted that, although in the claims a dependent claim refers to a specific combination with one or more other claims, other examples may also include a combination of a dependent claim with the subject matter of any other dependent or independent claim. Such combinations are expressly proposed herein, unless it is stated in individual cases that a specific combination is not intended. In addition, features of a claim should also be included in any other independent claim, even if the claim is not directly defined as a dependent claim of the other independent claim.

Claims

1. An apparatus (100) for measuring the heart rate of a target in a field of view of a radar sensor, the apparatus (100) comprising a processing circuit device (110), the processing circuit device (110) being configured to: Determining a frequency spectrum (200-1) corresponding to the time domain data (130) based on the received signal; determining a respiratory rate frequency based on a maximum amplitude (210-1) of the spectrum (200-1); Determining a plurality of segment frequency spectra based on the respiratory rate frequency, each segment frequency spectra corresponding to a corresponding segment of the time domain data (130); Determining the mean of the segment spectra to obtain an average spectrum (200-2); subtracting the average spectrum (200-2) from the spectrum (200-1) to obtain a difference spectrum (200-3); and The heart rate of the target is estimated based on the maximum amplitude (230-1) of the difference spectrum (200-3).

2. The apparatus (100) according to claim 1, wherein the processing circuit device (110) is configured to: applying a high pass filter for processing of the time domain data (130), Calculate the average amplitude of the high-pass filtered time domain data to obtain the noise level average, and If the noise level average is lower than a pre-specified noise level threshold, the frequency spectrum is determined (200-1).

3. The apparatus (100) according to claim 1 or claim 2, wherein the processing circuit device (110) is configured to: Calculating the average value of the amplitude values ​​of the time domain data (130), subtracting the mean value from the amplitude values ​​of the time domain data (130) to generate mean-removed time domain data, and The frequency spectrum is determined for the de-averaged time domain data.

4. The apparatus (100) according to any one of the preceding claims, wherein the processing circuit arrangement (110) is configured to: performing one or more range Fourier transforms of the time domain data (130) to obtain one or more range buffers, each range buffer being divided into a plurality of range bins, For each range buffer, identifying the range bin associated with the maximum amplitude (422) to determine a range-time data set (430), and The frequency spectrum is determined for the range-time data set (430) corresponding to the identified range bin (422).

5. The apparatus (100) according to claim 4, wherein the processing circuit device (110) is configured to: checking whether the peak-to-peak value for the distance-to-time data set (430) is within pre-specified limits based on an average of a plurality of previous peak-to-peak values ​​in the distance-to-time data set (430), and If the peak-to-peak value is within its pre-specified limits, the frequency spectrum is determined for the distance-time data set (430).

6. The apparatus (100) according to claim 4 or claim 5, wherein the processing circuit device (110) is configured to: determining a plurality of derivative values ​​corresponding to the rate of change of the distance at corresponding times of the distance-time data set (430), determining the variance of the plurality of derivative values, checking whether each of the derivative values ​​is within a threshold determined by the variance, and If each derivative value is within the threshold, the frequency spectrum (200-1) for the distance-time data set (430) is determined.

7. An apparatus (100) according to any one of the preceding claims, wherein the processing circuit device (110) is configured to determine the frequency spectrum (200-1) for the time domain data (130) and subsequently apply a finite impulse response filter to the amplitude values ​​of the frequency spectrum (200-1).

8. An apparatus (100) according to any one of the preceding claims, wherein the processing circuit device (110) is configured to determine the respiratory rate frequency by detecting a first frequency and detecting a second frequency, wherein the first frequency corresponds to a maximum amplitude (210-1) value of the spectrum (200-1), and the second frequency corresponds to a local maximum (210-2) of the amplitude value in the spectrum (200-1) within a pre-specified frequency range starting from twice the first frequency.

9. The apparatus (100) of claim 8, wherein the processing circuitry (110) is configured to verify that a maximum amplitude (220-1) value of the average spectrum (200-2) corresponds to a frequency within a pre-specified frequency range from the respiratory rate frequency.

10. The apparatus (100) of claim 9, wherein for a case where the maximum amplitude (220-1) value of the average spectrum (200-2) corresponds to a frequency that is not within the pre-specified frequency range from the respiratory rate frequency, the processing circuit means (110) is configured to: determining a new respiratory rate frequency based on adjacent local maxima of the amplitude values ​​in the spectrum (200-1), Obtain a new average spectrum based on the new respiratory rate frequency, It is verified that the maximum amplitude value of the new average spectrum corresponds to a frequency within the pre-specified frequency range from the new breathing rate frequency.

11. The apparatus (100) according to any one of the preceding claims, wherein the processing circuit arrangement (110) is configured to: performing a first correlation operation on the spectrum (200-1) to determine a correlated spectrum (730-3a), and performing a second correlation operation on the average spectrum (200-2) to determine a correlated average spectrum (730-3b), wherein each correlation operation is between an original version (730-1) and a windowed version (730-2) of the corresponding spectrum, wherein the corresponding windowed version (730-2) is changed by applying a first Gaussian window (732-1a; 732-1b) centered at a first frequency corresponding to the maximum amplitude (210-1; 220-1) of the spectrum, applying a second Gaussian window (732-2a; 732-2b) centered at a second frequency twice the first frequency, applying a third Gaussian window (732-3a; 732-3b) centered at a third frequency three times the first frequency, and invalidating all amplitude values ​​outside the three Gaussian windows; and The correlation average spectrum (730-3b) is subtracted from the correlation spectrum (730-3a) to obtain the difference spectrum (200-3).

12. The apparatus (100) of any one of the preceding claims, wherein the processing circuitry (110) is configured to perform multiple heart rate estimations, wherein each estimation is applied to an iteration of a recursive model for obtaining an accurate heart rate estimate.

13. A radar system (900), comprising: The device (910) according to any one of claims 1 to 12; as well as The radar sensor (920) is configured to generate the received signal.

14. A method (1000) for measuring the heart rate of a target in a field of view of a radar sensor, the method comprising: determining a frequency spectrum corresponding to the time domain data based on the received signal (1010); determining a respiratory rate frequency based on a maximum amplitude of the frequency spectrum (1020); Determining a plurality of segment frequency spectra based on the respiratory rate frequency (1030), each segment frequency spectra corresponding to a respective segment of the time domain data; determining a mean of the frequency spectra of the segments (1040) to obtain an average spectrum; subtracting (1050) the average spectrum from the spectrum to obtain a difference spectrum; and The heart rate of the target is estimated (1060) based on the maximum amplitude of the difference spectrum.

15. A program having a program code for executing the method according to claim 14 when the program is executed on a processor or programmable hardware.

Images