A radar-based life detection method and apparatus

Abstract

This invention discloses a radar-based life detection method and apparatus. The method includes: radar detection of a target, determining the target's range and angle information; preprocessing signals within the target's range cell to output a target signal; acquiring the phase information of the target signal, performing phase unwrapping on the target signal to obtain a life signal; separating the life signal to obtain a respiratory signal and a heartbeat signal; obtaining the frequencies of the heartbeat signal and the respiratory signal using time-domain and frequency-domain methods; and determining respiratory frequency weighting coefficients and heartbeat frequency weighting coefficients based on the numerical differences between the respiratory frequency and heartbeat frequency under the time-domain and frequency-domain methods, respectively, to obtain respiratory frequency detection values ​​and heartbeat frequency detection values. This method effectively reduces the drawbacks of both time-domain and frequency-domain methods, improving the accuracy of detecting vital sign data.

Description

Technical Field

[0001] This invention relates to the field of radar, and more particularly to a radar-based method and apparatus for life detection. Background Technology

[0002] Currently, the method of detecting breathing and heartbeat using millimeter-wave radar is to output the maximum peak value of the spectrum within the breathing and heartbeat frequency band. However, the detection accuracy of breathing and heartbeat frequencies is extremely susceptible to being affected by body movement.

[0003] However, if respiratory and heart rate are calculated using time-domain methods, short-term strong disturbances will only have a minor impact on the results. But time-domain methods for calculating respiratory and heart rate are not sensitive enough to changes in respiratory and heart rate, and their accuracy is still not high enough.

[0004] Therefore, there is a need to provide a radar-based life detection method and device to improve the accuracy of detecting respiratory and heart rate. Summary of the Invention

[0005] The purpose of this invention is to provide a radar-based life detection method and device to improve the accuracy of detecting respiratory and heart rate.

[0006] To achieve the above objectives, the present invention provides a radar-based life detection method, comprising:

[0007] The radar detects the target and determines its distance and angle information.

[0008] Based on the distance and angle information, perform fast Fourier transform on the signal of the target's distance unit in each channel, and perform multi-channel digital beamforming to output the target signal;

[0009] Extract the phase information of the target signal, perform phase unwrapping on the target signal, and obtain the life signal;

[0010] The vital signs are separated to obtain respiratory and heartbeat signals;

[0011] The first heartbeat frequency of the heartbeat signal is obtained by a time-domain method, and the second heartbeat frequency of the heartbeat signal is obtained by a frequency method.

[0012] Based on the breathing signal, determine whether the target is in a breath-holding state; if not, obtain the first breathing frequency of the breathing signal using a time-domain method and obtain the second breathing frequency of the breathing signal using a frequency-domain method.

[0013] Based on the difference between the first respiratory rate and the second respiratory rate, a respiratory rate weighting coefficient is determined, and a respiratory rate detection value is obtained; based on the difference between the first heart rate and the second heart rate, a heart rate weighting coefficient is determined, and a heart rate detection value is obtained.

[0014] Optionally, the angle information includes pitch angle information and horizontal angle information.

[0015] Optionally, a steering vector is constructed based on the pitch angle information and the horizontal angle information before digital beamforming.

[0016] Optionally, the radar is a millimeter-wave radar.

[0017] Optionally, a sliding window can be applied to one or more of the phase unwrapping, time-domain, and frequency-domain methods.

[0018] Optionally, the method for determining whether the detection target is in a breath-holding state includes: acquiring the time-domain change of the respiratory signal; if the time-domain change is less than a threshold, then the detection target is in a breath-holding state.

[0019] Optionally, the method for separating the respiratory signal and the heartbeat signal includes: bandpass filtering the vital signals using a respiratory filter and a heartbeat filter.

[0020] Optionally, the frequency ranges of the breathing filter and the heartbeat filter are determined based on the target's breathing frequency range and heartbeat frequency range, respectively; the frequency ranges of the breathing filter and the heartbeat filter may partially overlap or completely not overlap.

[0021] Optionally, the method for obtaining the first respiratory frequency and the first heartbeat frequency using a time-domain method includes: determining the heartbeat waveform period and the respiratory waveform period based on the heartbeat frequency range of the heartbeat signal and the respiratory frequency range of the respiratory signal, respectively; determining the effective peak points of the respiratory signal and the heartbeat signal; and obtaining the respiratory frequency of the respiratory signal and the heartbeat frequency of the heartbeat signal based on the effective peak points.

[0022] The present invention also provides a radar-based life detection device, comprising:

[0023] The detection module is used to acquire information about targets detected by the radar.

[0024] The signal processing module is used to acquire vital signs based on the information of the target detected by the radar, and to perform phase extraction and separation on the vital signs to acquire respiratory signals and heartbeat signals;

[0025] The signal determination module is used to determine whether the radar-detected target is in a breathless state;

[0026] The signal processing module is used to combine time-domain and frequency-domain methods to obtain the respiratory rate of the respiratory signal and the heart rate of the heartbeat signal.

[0027] The radar-based life detection method of this invention, after acquiring the target signal through millimeter-wave radar, first performs a series of processes on the target signal to obtain the vital signs, then separates the heartbeat and respiration signals from the vital signs, and then calculates the frequencies of the heartbeat and respiration signals using both time-domain and frequency-domain methods. Finally, it calculates the weighting coefficients of the heartbeat and respiration signals based on the frequency difference between the two methods, so that the frequency results of the time-domain method and the frequency-domain method can be combined to calculate a more accurate heartbeat and respiration frequency. This method effectively reduces the drawbacks of each of the time-domain and frequency-domain methods, and improves the accuracy of vital sign data detection. Attached Figure Description

[0028] Figure 1 This is a flowchart of a radar-based life detection method in a specific embodiment of the present invention;

[0029] Figure 2 This is a block diagram of a radar-based life detection device according to a specific embodiment of the present invention;

[0030] Figure 3 This is a data waveform diagram of radar-based life detection according to a specific embodiment of the present invention;

[0031] Figure 4 This is a data waveform diagram of radar-based life detection according to another specific embodiment of the present invention. Detailed Implementation

[0032] The radar-based life detection method and apparatus of the present invention will now be described in more detail with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. It should be understood that those skilled in the art can modify the invention described herein while still achieving its advantageous effects. Therefore, the following description should be understood as being of general knowledge to those skilled in the art and is not intended to limit the invention.

[0033] Please refer to Figure 1 , Figure 1 This is a flowchart of a radar-based life detection method according to a specific embodiment of the present invention.

[0034] The radar-based life detection method in this embodiment includes:

[0035] Step S101: The radar detects the target and determines the target's distance and angle information;

[0036] Step S102: Based on the distance information and angle information, perform fast Fourier transform on the signal of the distance unit of the target in each channel, and perform multi-channel digital beamforming to output the target signal;

[0037] Step S103: Extract the phase information of the target signal, perform phase unwrapping on the target signal, and obtain the life signal;

[0038] Step S104: Separate the vital signs and obtain the respiratory and heartbeat signals;

[0039] Step S105: Obtain the first heartbeat frequency of the heartbeat signal using a time-domain method, and obtain the second heartbeat frequency of the heartbeat signal using a frequency method;

[0040] Step S106: Determine whether the target is in a breath-holding state based on the breathing signal; if not, obtain the first breathing frequency of the breathing signal using a time-domain method and obtain the second breathing frequency of the breathing signal using a frequency-domain method.

[0041] Step S107: Determine the respiratory rate weighting coefficient based on the difference between the first respiratory rate and the second respiratory rate, and obtain the respiratory rate detection value; determine the heart rate weighting coefficient based on the difference between the first heart rate and the second heart rate, and obtain the heart rate detection value.

[0042] Specifically, in step S101, the radar transmits a signal to the target to be detected and receives the reflected signal from the target. Based on the transmitted and reflected signals, the distance and angle information of the target are determined. In this specific embodiment, the radar is a millimeter-wave radar, which operates in the millimeter-wave band and can efficiently distinguish the vital signs of a target.

[0043] Furthermore, the angle information includes pitch angle information and horizontal angle information.

[0044] In step S102, the target position is determined based on the range information. Within the range cell where the target is located, a range-dimensional Fast Fourier Transform is performed on the radar signal of the target in each channel of the range cell. Multi-channel digital beamforming is then performed in conjunction with the angle information, and finally, the target signal is output. Digital beamforming can effectively improve the target signal-to-noise ratio and enhance the effectiveness of the target signal.

[0045] Furthermore, in this specific embodiment, the angle information includes pitch angle information and horizontal angle information. Before performing digital beamforming, the angle information is first processed, that is, a guide vector is constructed based on the pitch angle information and horizontal angle information, thereby performing digital beamforming that combines the horizontal angle and pitch angle.

[0046] In step S103, the phase information in the target signal is calculated, and the true phase information of the target is obtained by phase unwrapping, thereby obtaining the life signal.

[0047] Furthermore, phase unwrapping uses a sliding window to store the phase, and the phase unwrapping result is stored in the last cell of the sliding window. In one specific embodiment, the sliding window length is 256.

[0048] Furthermore, the time-domain method, frequency-domain method, and other methods for processing signal data in this embodiment can also utilize sliding window storage and computation, thereby reducing computational complexity and improving signal processing efficiency.

[0049] In step S104, the vital signs in the sliding window are separated to obtain the respiratory and heartbeat signals. In this specific embodiment, the method for separating the respiratory and heartbeat signals is to perform bandpass filtering on the vital signs using a respiratory filter and a heartbeat filter.

[0050] Specifically, because respiratory and heartbeat signals have different frequency ranges, respiratory and heartbeat filters can be set to their respective frequency ranges. That is, the respiratory filter's frequency range is the same as the respiratory signal's frequency range, and the heartbeat filter's filtering frequency range is the same as the heartbeat signal's frequency range, thus acquiring both the respiratory and heartbeat signals. Furthermore, the frequency ranges of the respiratory and heartbeat signals can partially overlap or completely overlap.

[0051] For example, in one specific implementation, the frequency range of the respiratory signal is 0.1Hz-0.75Hz, and the frequency range of the heartbeat signal is 0.7Hz-2Hz; then the frequency range of the respiratory filter is set to 0.1Hz-0.75Hz, and the frequency range of the heartbeat filter is 0.7Hz-2Hz.

[0052] In steps S104-S106, it is determined whether the target is in a breath-holding state based on the breathing signal. Specifically, the phase difference between adjacent frames of the breathing signal is calculated, and the phase difference result is stored in a sliding window. Since each frame corresponds to one sliding window, the formula for calculating the phase difference is: The breathing time-domain value is output for each frame, where N is the length of the sliding window and is a natural number greater than 1. Then, the average phase difference value of the multiple phase differences stored within the sliding window is calculated using the following formula: Finally, the change in phase difference is calculated using the following formula: This refers to the time-domain variation of the respiratory signal.

[0053] By comparing the time-domain change with a set threshold, it is possible to determine whether the target is in a breath-holding state. For example, if the threshold is set to 2, when the time-domain change of the respiratory signal is less than 2, it means that the chest cavity changes caused by the target's breathing and heartbeat are very small, and therefore, the target can be judged to be in a breath-holding state.

[0054] If the target is holding its breath, the respiratory signal frequency is not acquired; only the heartbeat signal frequency is acquired. If the target is not holding its breath, the frequencies of both the heartbeat and respiratory signals are acquired. Specifically, the first respiratory frequency and first heartbeat frequency of the respiratory and heartbeat signals are acquired using a time-domain method; and the second respiratory frequency and second heartbeat frequency of the respiratory and heartbeat signals are acquired using a frequency-domain method.

[0055] Specifically, the method for obtaining the first respiratory frequency and the first heartbeat frequency using the time domain method is as follows: determine the heartbeat waveform period and the respiratory waveform period according to the heartbeat frequency range of the heartbeat signal and the respiratory frequency range of the respiratory signal, respectively; determine the peak points of the respiratory signal and the heartbeat signal, with the first peak point being the effective peak point; based on the heartbeat waveform period and the respiratory waveform period, starting from the first peak point, determine other effective peak points in sequence, that is, peak points whose time distance from the previous effective peak point is within the waveform period to which they belong are determined to be effective peak points.

[0056] For example, when the frequency range of the respiratory signal is 0.1-0.75, the waveform period of the respiratory signal is between 1.33s and 10s. To obtain the peak points of the respiratory signal, firstly, the first peak point of the respiratory signal is determined as the effective peak point. Starting from the first effective peak point, the peak points with a time interval between 1.33s and 10s from the first effective peak point are determined as the second effective peak points. Then, starting from the second effective peak point, the third effective peak point is determined according to the period time range. Then, starting from the third effective peak point, the fourth effective peak point is determined according to the period time range, and so on, until all effective peak points of the respiratory signal are obtained. Finally, the total number of effective peak points is counted, and the first respiratory frequency is calculated.

[0057] The method for obtaining the second respiratory frequency and the second heartbeat frequency using the frequency method is as follows: perform Fast Fourier Transform on the respiratory signal and the heartbeat signal respectively to obtain the respiratory spectrum and the heartbeat spectrum; find the maximum value on the respiratory spectrum and the heartbeat spectrum respectively to obtain the second respiratory frequency and the second heartbeat frequency.

[0058] In step S107, a respiratory rate weighting coefficient is determined based on the numerical difference between the first and second respiratory rates to obtain the respiratory rate detection value. In this embodiment, the respiratory rate weighting coefficient is determined as follows: if the numerical difference between the first and second respiratory rates exceeds 10 breaths / minute, the weight of the time-domain result is 1, and the weight of the frequency result is 0; if the numerical difference between the first and second respiratory rates does not exceed 10 breaths / minute, the weight of the time-domain result is 0.3, and the weight of the frequency result is 0.7. In other specific embodiments, different respiratory rate weighting coefficient calculation methods can be set according to specific objectives.

[0059] By combining the time-domain result coefficient and the frequency result coefficient in the respiratory rate weighting coefficient, the respiratory rate detection value can be obtained.

[0060] Based on the difference between the first and second heart rate frequencies, a weighting coefficient for the heart rate is determined to obtain the detected heart rate value. In this embodiment, if the difference between the first and second heart rate frequencies exceeds 15 beats per minute, the weight of the time-domain result is 1, and the weight of the frequency-domain result is 0; if the difference between the first and second heart rate frequencies does not exceed 15 beats per minute, the weight of the time-domain result is 0.5, and the weight of the frequency-domain result is 0.5. In other specific embodiments, different methods for calculating the heart rate weighting coefficient can be set according to the specific target.

[0061] By combining the time-domain and frequency coefficients in the heart rate weighting coefficients, the heart rate detection value can be obtained. Please refer to [reference needed]. Figure 3 and Figure 4 , Figure 3 This is a graph showing the data results of heart rate detection using only time-domain methods. Figure 4 This is a graph showing the data results calculated using time-domain and frequency-domain weighting in this embodiment. It is clear that... Figure 4 The error between the radar detection results and the reference value is much smaller than Figure 3 The radar-based life detection method in this embodiment greatly improves the accuracy of detecting vital signs by reducing the error between the radar detection results and the reference values.

[0062] The radar-based life detection method of this invention, after acquiring the target signal through millimeter-wave radar, first performs a series of processes on the target signal to obtain the vital signs, then separates the heartbeat and respiration signals from the vital signs, and then calculates the frequencies of the heartbeat and respiration signals using both time-domain and frequency-domain methods. Finally, it calculates the weighting coefficients of the heartbeat and respiration signals based on the frequency difference between the two methods, so that the frequency results of the time-domain method and the frequency-domain method can be combined to calculate a more accurate heartbeat and respiration frequency. This method effectively reduces the drawbacks of each of the time-domain and frequency-domain methods, and improves the accuracy of vital sign data detection.

[0063] Please refer to Figure 2 , Figure 2 This is a block diagram of a radar-based life detection device according to a specific embodiment of the present invention.

[0064] This embodiment provides a radar-based life detection device, including:

[0065] Detection module 201 is used to acquire information about targets detected by radar;

[0066] The signal processing module 202 is used to acquire life signals based on the information of the radar-detected target, and to perform phase extraction and separation on the life signals to acquire respiratory signals and heartbeat signals;

[0067] Signal determination module 203 is used to determine whether the radar-detected target is in a breathless state;

[0068] The signal processing module 204 is used to combine time-domain and frequency-domain methods to obtain the respiratory frequency of the respiratory signal and the heart rate of the heart rate of the heart rate signal.

[0069] Specifically, the detection module 201 is used for the radar to transmit signals to the target to be detected and to receive the reflected signals from the target, and to obtain target distance and angle information based on the radar signals.

[0070] The signal processing module 202 is used to perform fast Fourier transform on the signal of the target's distance unit according to the distance information and angle information, and to perform multi-channel digital beamforming to output the target signal; to extract the phase and unwrap the phase of the target signal to obtain the vital signal; and to separate the vital signal through a bandpass filter to obtain the heartbeat signal and the breathing signal.

[0071] The signal judgment module 203 is used to determine whether the target is in a breath-holding state based on the comparison result of the time-domain change of the respiratory signal and the threshold. If the target is in a breath-holding state, the respiratory rate is not calculated. If the target is not in a breath-holding state, the respiratory rate is calculated.

[0072] The signal processing module 204 is used to obtain the first heartbeat frequency of the heartbeat signal using a time-domain method and the second heartbeat frequency of the heartbeat signal using a frequency method; to obtain the first respiratory frequency of the respiratory signal using a time-domain method and the second respiratory frequency of the respiratory signal using a frequency-domain method; to determine a respiratory frequency weighting coefficient based on the numerical difference between the first and second respiratory frequencies and obtain a respiratory frequency detection value; and to determine a heartbeat frequency weighting coefficient based on the numerical difference between the first and second heartbeat frequencies and obtain a heartbeat frequency detection value.

[0073] The radar-based life detection device of this invention, through the cooperation of a detection module, a signal processing module, a signal judgment module, and a signal calculation module, acquires signal data by using radar to detect targets. After processing the signal data, it obtains separate respiratory and heartbeat signals. Combining time-domain and frequency-based methods for acquiring the frequencies of the respiratory and heartbeat signals, it detects the target's respiratory and heartbeat frequencies. This radar-based life detection device can more accurately detect the target's vital signs, with higher accuracy in detecting respiratory and heartbeat frequencies, effectively improving the efficacy of radar-based life detection devices.

[0074] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A radar-based life detection method, characterized in that, include: The radar detects the target and determines its distance and angle information. Based on the distance and angle information, perform fast Fourier transform on the signal of the target's distance unit in each channel, and perform multi-channel digital beamforming to output the target signal; Extract the phase information of the target signal, perform phase unwrapping on the target signal, and obtain the life signal; The vital signs are separated to obtain respiratory and heartbeat signals; The first heartbeat frequency of the heartbeat signal is obtained by a time-domain method, and the second heartbeat frequency of the heartbeat signal is obtained by a frequency method. Determine whether the target is in a breath-holding state based on the breathing signal; If not, the first respiratory frequency of the respiratory signal is obtained by the time domain method, and the second respiratory frequency of the respiratory signal is obtained by the frequency domain method. Based on the numerical difference between the first respiratory rate and the second respiratory rate, a respiratory rate weighting coefficient is determined. The respiratory rate detection value is obtained by combining the time domain result coefficient and the frequency result coefficient in the respiratory rate weighting coefficient with the time domain and the frequency domain. Based on the numerical difference between the first and second heart rate frequencies, a weighting coefficient for the heart rate is determined. The heart rate detection value is then obtained by combining the time-domain result coefficient and the frequency result coefficient in the weighting coefficient with the time-domain and frequency-domain results.

2. The radar-based life detection method according to claim 1, characterized in that, The angle information includes pitch angle information and horizontal angle information.

3. The radar-based life detection method according to claim 2, characterized in that, Before digital beamforming, a steering vector is constructed based on the pitch angle information and the horizontal angle information.

4. The radar-based life detection method according to claim 1, characterized in that, The radar in question is a millimeter-wave radar.

5. The radar-based life detection method according to claim 1, characterized in that, Apply sliding windows to one or more of the phase unwrapping, time-domain, and frequency-domain methods.

6. The radar-based life detection method according to claim 1, characterized in that, The method for determining whether the target is in a breath-holding state includes: acquiring the time-domain change of the respiratory signal; if the time-domain change is less than a threshold, then the target is in a breath-holding state.

7. The radar-based life detection method according to claim 1, characterized in that, The method for separating the vital signals and obtaining respiratory and heartbeat signals includes: performing bandpass filtering on the vital signals using a respiratory filter and a heartbeat filter.

8. The radar-based life detection method according to claim 7, characterized in that, The frequency ranges of the breathing filter and the heartbeat filter are determined based on the target's breathing frequency range and heartbeat frequency range, respectively; the frequency ranges of the breathing filter and the heartbeat filter may partially overlap or not overlap at all.

9. The radar-based life detection method according to claim 1, characterized in that, The method for obtaining the first respiratory frequency and the first heartbeat frequency using a time-domain approach includes: determining the heartbeat waveform period and the respiratory waveform period based on the heartbeat frequency range of the heartbeat signal and the respiratory frequency range of the respiratory signal, respectively; determining the effective peak points of the respiratory signal and the heartbeat signal; and obtaining the respiratory frequency of the respiratory signal and the heartbeat frequency of the heartbeat signal based on the effective peak points.

10. A radar-based life detection device for implementing the radar-based life detection method as described in any one of claims 1 to 9, characterized in that, include: The detection module is used to acquire information about targets detected by the radar. The signal processing module is used to acquire vital signs based on the information of the target detected by the radar, and to perform phase extraction and separation on the vital signs to acquire respiratory signals and heartbeat signals; The signal determination module is used to determine whether the radar-detected target is in a breathless state; The signal processing module is used to combine time-domain and frequency-domain methods to obtain the respiratory rate of the respiratory signal and the heart rate of the heartbeat signal.

Images