Signal processing method and system

Abstract

The invention discloses a signal processing method and system, and the method comprises the steps: continuously transmitting a pulse signal which is used for detecting the breath of a target user and the heartbeat of the target user; receiving an echo signal of the pulse signal; based on the echo signal processing, obtaining a first respiratory frequency for representing the respiration of the target user, a first heartbeat frequency for representing the heartbeat of the target user, and a confidence coefficient for representing the first respiratory frequency and the first heartbeat frequency for representing the credibility of the respiration and the heartbeat of the target user; and at least outputting a second respiratory rate and a second heartbeat rate based on the first respiratory rate, the first heartbeat rate and the confidence.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and more specifically to a signal processing method and system. Background Technology

[0002] To obtain vital signs signals (such as heartbeat and respiration) from a human body, it is usually necessary to use professional monitoring instruments to collect the signals. For example, an electrocardiogram (ECG) machine is needed to collect the human heartbeat signal. However, these contact-based professional monitoring instruments have certain limitations, which reduce the convenience and efficiency of signal detection. Summary of the Invention

[0003] In view of the above, this application provides the following technical solution:

[0004] A signal processing method, comprising:

[0005] Continuously transmit pulse signals, which are used to detect the target user's breathing and the target user's heartbeat;

[0006] Receive the echo signal of the pulse signal;

[0007] Based on the echo signal processing, a first respiratory rate for characterizing the breathing of the target user and a first heart rate for characterizing the heartbeat of the target user are obtained, and a confidence level for characterizing the first respiratory rate and the first heart rate is used to characterize the reliability of the breathing and heartbeat of the target user.

[0008] Based on the first respiratory rate, the first heart rate, and the confidence level, at least the second respiratory rate and the second heart rate are output.

[0009] Optionally, the step of obtaining a first respiratory rate to characterize the target user's breathing and a first heart rate to characterize the target user's heartbeat based on the echo signal processing, and a confidence level to characterize the first respiratory rate and the first heart rate as a representation of the reliability of the target user's breathing and heartbeat, includes:

[0010] Based on the echo signal, the respiratory signal waveform and the heartbeat signal waveform are obtained respectively;

[0011] Time-domain analysis was performed on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first respiratory frequency and the first heartbeat frequency;

[0012] The time-domain peak values ​​of the respiratory signal waveform and the heartbeat signal waveform are filtered respectively to obtain the first effective peak value corresponding to the respiratory signal waveform and the second effective peak value corresponding to the heartbeat signal waveform;

[0013] The respiratory reference frequency and the heart rate reference frequency are determined based on the first time interval characteristics of the first effective peak value and the second time interval characteristics of the second effective peak value, respectively.

[0014] A confidence level is determined based on the deviation between the first respiratory rate and the respiratory reference rate, and the deviation between the first heart rate and the heart rate reference rate. The confidence level is used to characterize the reliability of the first respiratory rate and the first heart rate as characterizes the breathing and heartbeat of the target user.

[0015] Optionally, obtaining the respiratory signal waveform and the heartbeat signal waveform based on the echo signal includes:

[0016] Based on the echo signal, a two-dimensional signal array containing a distance dimension and a time series dimension is generated; the distance dimension is divided into multiple distance units based on a fixed distance interval, and each distance unit corresponds to a specific distance interval between the signal source and the target user; the time series dimension consists of continuous time frames, and each time frame corresponds to a complete pulse transmission and reception cycle.

[0017] For each distance unit, the phase change sequence of the distance unit in the time series dimension is extracted as the first feature waveform;

[0018] The first characteristic waveform is processed by a filter corresponding to the first frequency of the breathing characteristic to obtain the breathing signal waveform;

[0019] For each time frame, the amplitude distribution sequence of the time frame in the distance dimension is extracted as the second feature waveform;

[0020] The second characteristic waveform is processed by a filter corresponding to the second frequency of the heartbeat characteristic to obtain the heartbeat signal waveform.

[0021] Optionally, the step of performing time-domain analysis on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first respiratory frequency and the first heartbeat frequency includes:

[0022] Perform time-domain analysis on the respiratory signal waveform to obtain all peak points in the respiratory signal waveform, and calculate the first time interval between adjacent peak points;

[0023] The main respiratory cycle is determined based on the frequency of occurrence of the first time interval;

[0024] The dominant frequency of the respiratory frequency domain is determined based on the frequency components in the first spectrum diagram;

[0025] Based on the main respiratory cycle, a first respiratory rate is determined;

[0026] Perform time-domain analysis on the heartbeat signal waveform to obtain all peak points in the heartbeat signal waveform, and calculate the second time interval between adjacent peak points;

[0027] The main heartbeat cycle is determined based on the frequency of occurrence of the second time interval;

[0028] The first heartbeat frequency is determined based on the main heartbeat cycle.

[0029] Optionally, determining the first respiratory rate based on the master respiratory cycle includes:

[0030] Convert the main respiratory cycle into a first frequency;

[0031] Frequency domain analysis was performed on the respiratory signal waveform to obtain the first spectrum.

[0032] Based on the first spectrum diagram, determine the dominant frequency of the respiratory signal waveform in the respiratory frequency domain;

[0033] The first respiratory frequency is determined based on the first frequency and the dominant frequency of the respiratory frequency domain.

[0034] Optionally, determining the confidence level based on the deviation between the first respiratory rate and the respiratory reference rate, and the deviation between the first heart rate and the heart rate reference rate, includes:

[0035] Calculate the first deviation between the first respiratory rate and the respiratory rate reference rate, and the second deviation between the first heart rate and the heart rate reference rate, respectively;

[0036] Count the first number of the first effective peak and the second number of the second effective peak;

[0037] The confidence level is determined based on the first deviation, the second deviation, the first quantity, and the second quantity.

[0038] Optionally, the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level includes:

[0039] Based on the confidence level of each distance unit corresponding to the echo signal, an effective distance unit is determined; wherein, each distance unit corresponds to a specific distance interval between the signal source and the target user;

[0040] The first respiratory rate and the first heart rate corresponding to the effective distance unit are respectively determined as the second respiratory rate and the second heart rate;

[0041] Output the second respiratory rate and the second heart rate.

[0042] Optionally, the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level includes:

[0043] If the confidence level is less than a preset threshold, the previously output respiratory rate and heart rate are determined as the second respiratory rate and the second heart rate, and the second respiratory rate and the second heart rate are output.

[0044] If the confidence level is not less than the preset threshold, the first respiratory rate is determined as the second respiratory rate, and the first heart rate is determined as the second heart rate; the second respiratory rate and the second heart rate are output.

[0045] Optionally, the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level includes:

[0046] The first respiratory rate is determined as the second respiratory rate, and the first heart rate is determined as the second heart rate;

[0047] The confidence level is converted into perceptible error information;

[0048] Output the second respiratory rate, the second heart rate, and the perceptible error information.

[0049] A signal processing system includes: a signal device and a processing device;

[0050] The signaling device is used to continuously transmit pulse signals, which are used to detect the breathing and heartbeat of the target user; and to receive the echo signals of the pulse signals.

[0051] The processing device is configured to obtain, based on the echo signal processing, a first respiratory frequency characterizing the breathing of the target user and a first heart rate characterizing the heartbeat of the target user, and a confidence level characterizing the first respiratory frequency and the first heart rate as a reliability of characterizing the breathing and heartbeat of the target user; and to output at least a second respiratory frequency and a second heart rate based on the first respiratory frequency, the first heart rate and the confidence level. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0053] Figure 1 A schematic flowchart of a signal processing method provided in an embodiment of this application;

[0054] Figure 2 A flowchart illustrating a method for determining confidence level provided in an embodiment of this application;

[0055] Figure 3 A schematic diagram of an echo signal provided in an embodiment of this application;

[0056] Figure 4 A multi-channel time-domain waveform diagram of the real components of a UWB signal provided in an embodiment of this application;

[0057] Figure 5 This application provides a multi-channel time-domain phase diagram of the imaginary component of a UWB signal.

[0058] Figure 6 This is a schematic diagram of a filtered respiratory signal waveform provided in an embodiment of this application;

[0059] Figure 7 This is a schematic diagram of a filtered heartbeat signal waveform provided in an embodiment of this application;

[0060] Figure 8 A schematic diagram showing the comparison between raw data and a smoothed curve provided in this embodiment;

[0061] Figure 9 This application provides a schematic diagram of signal processing for an application scenario.

[0062] Figure 10 This is a schematic diagram of the structure of a signal processing system provided in an embodiment of this application. Detailed Implementation

[0063] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0064] The terms "first" and "second," etc., used in this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units may include steps or units not listed, but may include steps or units not listed.

[0065] This application provides a signal processing method that can be applied to non-contact physiological signal monitoring of users, enabling the acquisition of the user's respiratory rate and heart rate. For example, it can be applied to home health monitoring, sleep state analysis, and medical care, accurately detecting respiratory and heart rates without contact with the user by continuously emitting pulse signals. This improves user comfort and the flexibility of usage scenarios. Furthermore, through time-domain analysis and confidence verification during signal processing, analysis errors are reduced, improving the accuracy, reliability, and stability of physiological signal detection.

[0066] See Figure 1 The diagram illustrates a flowchart of a signal processing method provided in an embodiment of this application, which may include the following steps:

[0067] S101, continuously transmit pulse signals.

[0068] Pulse signals are used to detect the breathing and heartbeat of a target user. These pulse signals can be continuously emitted by a signal transmitting module in a signal device. For example, the signal device can be based on a UWB (Ultra Wide Band) radar system. UWB-based human breathing and heartbeat signal detection can detect the breathing and heartbeat signals of a target user within a certain area, through certain media (such as walls, clothing, etc.), and without physical contact. For instance, the signal transmitting module can use nanosecond-level UWB pulses (e.g., a center frequency of 7 GHz and a pulse width of 1.5 ns) for continuous signal transmission. Alternatively, a frequency-modulated continuous wave (FMCW) radar or pulse Doppler radar can be used as the signal source, emitting pulse signals in the frequency range of 24 GHz to 77 GHz (this frequency band can effectively penetrate clothing and is harmless to the human body). The signal repetition frequency is set to 100 Hz to 1 kHz to ensure the capture of low-frequency movements of breathing (0.2-0.34 Hz) and heartbeat (0.8-2.5 Hz).

[0069] S102, Receive the echo signal of the pulse signal.

[0070] The echo signal of the pulse signal is received by the signal receiving module in the signal device, for example, by a receiving antenna array paired with the transmitting antenna. Typically, the echo signal is converted to an intermediate frequency signal by a mixer, then its signal strength is enhanced by a low-noise amplifier, and finally converted to a digital signal by an analog-to-digital converter. The sampling rate needs to be at least 1kHz to preserve the high-frequency details of the echo signal for subsequent processing.

[0071] S103. Based on echo signal processing, obtain a first respiratory frequency for characterizing the target user's breathing and a first heart rate for characterizing the target user's heartbeat, and a confidence level for characterizing the first respiratory frequency and the first heart rate as a reliable representation of the target user's breathing and heartbeat.

[0072] S104. Based on the first respiratory rate, the first heart rate, and the confidence level, output at least the second respiratory rate and the second heart rate.

[0073] After obtaining the echo signal, since there may be some interference signals in the echo signal, the echo signal can be denoised first to enhance the effective signal components, and then the pre-processed echo signal can be processed.

[0074] In this embodiment, the respiratory rate and heart rate of the target user can be synchronously detected. Therefore, to improve the accuracy of signal processing, it is necessary to separate the respiratory and heart rate signals from the echo signal based on the differences in their signal characteristics. Then, the separated respiratory and heart rate signals can be analyzed in the time domain (e.g., period detection) to obtain the first respiratory rate and the first heart rate, or they can be combined with frequency domain analysis (e.g., spectrum extraction) to obtain a comprehensive first respiratory rate and the first heart rate. To verify the reliability of the first respiratory rate and the first heart rate in representing the target user's breathing and heart rate, a confidence level is also calculated in this embodiment. That is, by comparing the consistency of the respiratory and heart rate results obtained by different analysis methods, and combining this with the stability of the signal waveform (e.g., determining the stability of the signal waveform through the peak patterns reflected in the signal waveform), the confidence level of the frequency representation results is calculated.

[0075] The confidence level is determined by comparing the deviations between the calculated respiratory and heart rates and the actual feedback frequencies of the signal waveforms. This provides a quantitative reliability assessment standard for physiological signal frequency detection, directly reflecting the reliability of the current detection results. For example, high confidence indicates consistency between the results of the two analysis dimensions and can be directly used as valid data output; low confidence indicates possible noise interference or signal anomalies, leading to deviations in the detection results. This approach not only improves the robustness of the output results and avoids the direct application of data with large errors, but also provides a clear decision-making basis for subsequent processing. For instance, in medical monitoring scenarios, confidence can help determine whether manual verification is needed; in home monitoring, error prompts can help users interpret the data reasonably. Ultimately, by quantifying "uncertainty," confidence calculation balances the real-time nature and accuracy of detection, ensuring that non-contact physiological signal monitoring remains practical in complex environments.

[0076] In step S104, after obtaining the confidence level, the respiratory rate and heart rate to be output are determined based on the confidence level. For example, if the calculated confidence level is high, the first respiratory rate and the first heart rate can be determined as the second respiratory rate and the second heart rate to be output. If the currently calculated confidence level is low, the respiratory rate and heart rate can be recalculated. If the new calculation result and the corresponding confidence level meet the requirement of high confidence, the new calculation result is output as the second respiratory rate and the second heart rate. Correspondingly, when the first respiratory rate is used as the output second respiratory rate and the first heart rate is used as the output second heart rate, the confidence level can also be output simultaneously. In order to make the reliability of the current detection clear to the user, the confidence level can also be converted into features that represent the detection error in a way that is easy for the user to understand, such as prompt sounds, highlighted colors, etc., so that the user can understand the reliability of the currently output heart rate and respiratory rate.

[0077] For example, in a home sleep monitoring scenario, a signal device can be placed above the bedside table to continuously transmit pulse signals to the user. After receiving the echo reflected from the chest cavity, the device first filters and separates the breathing and heartbeat signals. For instance, the breathing signal appears as a slow fluctuation with a period of approximately 3 seconds, while the heartbeat signal appears as a rapid vibration with a period of approximately 0.8 seconds. Analysis yields a first respiratory rate of 19 breaths / minute and a first heartbeat rate of 75 beats / minute. Since the deviation between the two analysis methods is less than 5%, the calculated confidence level is 0.89, meeting the high confidence requirement. The system determines the results to be reliable and directly outputs them as the second respiratory rate and second heartbeat rate to the accompanying application for recording.

[0078] This application provides a signal processing method that continuously transmits pulse signals and receives the echo signals of the pulse signals. The echo signals are processed to obtain a first respiratory rate representing the breathing of a target user and a first heart rate representing the heartbeat of the target user. A confidence level is also obtained, which represents the reliability of the current detection calculation. Through comprehensive analysis and confidence level evaluation mechanism, the influence of environmental interference on the measurement results is effectively reduced, and the accuracy and reliability of respiratory rate and heart rate detection are improved.

[0079] The following describes a signal processing method provided in the embodiments of this application, in conjunction with a practical application scenario.

[0080] After obtaining the echo signal, it is necessary to extract the signal waveforms of respiration and heartbeat to provide a data basis for subsequent calculations of respiratory and heart rate. In the embodiments of this application, the process of obtaining a first respiratory rate characterizing the target user's respiration and a first heart rate characterizing the target user's heartbeat based on echo signal processing, and a confidence level characterizing the first respiratory rate and the first heart rate as a representation of the reliability of the target user's respiration and heartbeat, may include... Figure 2 The relevant steps are shown in Figure 2, which illustrates a flowchart of a method for determining confidence levels. The corresponding steps are as follows:

[0081] S1031. Based on the echo signal, obtain the respiratory signal waveform and the heartbeat signal waveform respectively.

[0082] S1032. Perform time-domain analysis on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first respiratory frequency and the first heartbeat frequency.

[0083] S1033. Perform time-domain peak filtering on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first effective peak value corresponding to the respiratory signal waveform and the second effective peak value corresponding to the heartbeat signal waveform.

[0084] S1034. Determine the respiratory reference frequency and the heart rate reference frequency based on the first time interval characteristics of the first effective peak value and the second time interval characteristics of the second effective peak value, respectively.

[0085] S1035. Determine the confidence level based on the deviation between the first respiratory rate and the respiratory reference rate, and the deviation between the first heart rate and the heart rate reference rate.

[0086] The confidence level is used to characterize the first respiratory rate, and the first heart rate is used to characterize the credibility of the target user's breathing and heartbeat.

[0087] In step S1031, based on the time-domain variation characteristics of the echo signal, the respiratory signal waveform and the heartbeat signal waveform can be separated through targeted filtering. For example, the respiratory signal waveform is typically extracted using a bandpass filter that retains the 0.2-0.34Hz frequency components, reflecting the slow fluctuations of the chest cavity. The heartbeat signal waveform is extracted using a bandpass filter that retains the 0.8-2.5Hz frequency components, reflecting the rapid and minute vibrations of the chest cavity.

[0088] In some embodiments of this application, the process of obtaining respiratory signal waveforms and heartbeat signal waveforms based on echo signals may include the following steps: Based on the echo signals, a two-dimensional signal array containing a distance dimension and a time series dimension is generated; wherein, the distance dimension is divided into multiple distance units based on fixed distance intervals, each distance unit corresponding to a specific distance interval between the signal source and the target user; the time series dimension consists of continuous time frames, each time frame corresponding to a complete pulse transmission and reception cycle. Then, for each distance unit, the phase change sequence of the distance unit in the time series dimension is extracted as a first feature waveform; the first feature waveform is processed according to a filter of a first frequency corresponding to the respiratory feature to obtain the respiratory signal waveform; for each time frame, the amplitude distribution sequence of the time frame in the distance dimension is extracted as a second feature waveform; the second feature waveform is processed according to a filter of a second frequency corresponding to the heartbeat feature to obtain the heartbeat signal waveform.

[0089] In this implementation, to preserve the spatial distribution and temporal variation characteristics of the echo signal, the echo signal can be converted into a two-dimensional signal array. This two-dimensional signal array includes a distance dimension and a time series dimension. The distance dimension is divided into multiple distance units based on fixed distance intervals, with each distance unit corresponding to a specific distance region between the signal source and the target user. For example, the detection range of 0 to 5 meters can be divided into 100 distance units at 0.05 meters. This is because the movement of the human chest cavity is spatially localized (only the target user's location has obvious reflection), and the signal source can be located through distance units, eliminating interference from other areas. The time series dimension of the two-dimensional signal array is composed of continuous time frames, with each time frame corresponding to a complete pulse transmission and reception cycle. For example, 10 seconds of data can be collected at 0.01-second intervals to form 1000 time frames. This is done because breathing and heartbeat are dynamic signals that change over time. High temporal resolution (0.01 seconds) can capture the rapid vibrations of the heartbeat (e.g., a cycle of about 0.8 seconds), while a 0-second duration is sufficient to contain multiple respiratory cycles (about 3-5 seconds), ensuring the reliability of cycle analysis.

[0090] The displacement changes caused by respiration lead to phase changes in the echo signal. Therefore, the phase change of each distance unit in the time series dimension can be extracted as the first feature model. The minute movements caused by heartbeats lead to amplitude changes in the echo signal. Therefore, the amplitude distribution sequence of each time frame in the distance dimension can be extracted as the second feature waveform. Then, based on the respiratory characteristics, the first feature waveform is processed using a corresponding filter to obtain the respiratory signal waveform. For example, after filtering at 0.2-0.34Hz, the waveforms of the three distance units with the strongest signals are averaged to further suppress noise, while spatial correlation (i.e., the strongest signal in the distance unit area where the target user is located) is utilized to improve the signal-to-noise ratio of the respiratory waveform. Correspondingly, the second feature waveform can be processed using a filter corresponding to the second frequency of the heartbeat characteristics to obtain the heartbeat signal waveform. For example, after filtering the second feature waveform at 0.8-2.5Hz, a moving average (e.g., a moving window of 5 points) is applied to the feature value sequence to smooth high-frequency noise while preserving the rapid changes in heartbeat characteristics.

[0091] In step S1032, time-domain analysis can be performed on the respiratory signal waveform and the heartbeat signal waveform to obtain the first respiratory frequency and the first heartbeat frequency. The time-domain analysis determines the main period by detecting the peak interval of the signal (such as the fluctuation period of the respiratory waveform), reflecting the temporal continuity of the signal.

[0092] In some implementations, the process of performing time-domain analysis on the respiratory signal waveform and the heartbeat signal waveform to obtain the first respiratory frequency and the first heartbeat frequency may include: performing time-domain analysis on the respiratory signal waveform to obtain all peak points in the respiratory signal waveform, and calculating the first time interval between adjacent peak points; determining the main respiratory cycle based on the frequency of occurrence of the first time interval; and determining the first respiratory frequency based on the main respiratory cycle. Similarly, the above processing is performed on the heartbeat signal waveform, that is, performing time-domain analysis on the heartbeat signal waveform to obtain all peak points in the heartbeat signal waveform, and calculating the second time interval between adjacent peak points; determining the main heartbeat cycle based on the frequency of occurrence of the second time interval; and determining the first heartbeat frequency based on the second frequency corresponding to the main heartbeat cycle.

[0093] In this implementation, time-domain analysis extracts the actual periodic characteristics of the signal from its temporal variations. The amplitude of a user's physiological signals changes with the environment; fixed thresholds are prone to missed or false detections. Adaptive thresholds can dynamically match signal intensity, accurately identifying the peaks of respiration and heartbeat. Therefore, the adaptive threshold method can be used to detect signal waveforms and obtain the corresponding peak points. The threshold in the adaptive threshold can be set based on the baseline and standard deviation, e.g., threshold = baseline + 1.8 * standard deviation. The baseline represents the benchmark or average level of the signal under normal conditions. For example, in respiratory and heartbeat signal processing, the baseline might be the average value of the signal over a period of time (such as a stable period without abnormal interference), representing the normal benchmark of human physiological activity. The standard deviation is a statistical measure of the dispersion of signal data, reflecting the range of signal fluctuations near the baseline. For example, normal physiological signals (such as respiratory rate and heartbeat amplitude) are not completely constant and will fluctuate around the baseline. The standard deviation quantifies the magnitude of this fluctuation; a larger standard deviation indicates more drastic fluctuations in the signal under normal conditions, and vice versa. After obtaining all peak points in the signal waveforms (including respiratory and heartbeat waveforms), a timestamp (in seconds) for each peak is recorded. Calculating the time difference between adjacent peaks yields the peak interval sequence of the respiratory signal (i.e., the first time interval) and the peak interval sequence of the heartbeat signal (i.e., the second time interval). These time intervals directly reflect the periodic characteristics of the corresponding physiological signals. A grouping tolerance of 0.1 (i.e., a 10% error range) can be set to classify time intervals with differences within a reasonable range as having the same periodic characteristic. For example, a 3.0-second interval in the respiratory signal will include intervals in the range of 2.7-3.3 seconds in the same group, preserving the natural fluctuations of the period while avoiding the inclusion of abnormal intervals caused by interference.

[0094] To eliminate spurious peak intervals caused by noise, a minimum group size of min_group_size can be set to min_group_size=3 to ensure the reliability of the statistical results. Groups containing ≥3 intervals are selected as valid groups from all groups. For example, a group in the respiratory signal containing 5 intervals (3.1s, 3.2s, 3.0s, 3.3s, 3.2s) is considered a valid group because the number is ≥3. The average interval of each valid group is calculated as the time interval for that group. The valid group with the most intervals (the most common interval group) is selected, and its corresponding time interval is the "major period" (respiratory major period T_resp or heartbeat major period T_heart). The major period can then be converted to the corresponding frequency, such as frequency = 60 / major period.

[0095] Furthermore, to improve the accuracy of determining the first respiratory frequency and the first heartbeat frequency, frequency domain analysis can be performed on the respiratory signal waveform and the heartbeat signal waveform, and the results of the frequency domain analysis can be used as a reference. Correspondingly, in one embodiment of this application, the process of determining the first respiratory frequency based on the main respiratory cycle may further include: converting the main respiratory cycle into a first frequency; performing frequency domain analysis on the respiratory signal waveform to obtain a first spectrum; determining the dominant respiratory frequency of the respiratory signal waveform based on the first spectrum; and determining the first respiratory frequency based on the first frequency and the dominant respiratory frequency. Similarly, the process of determining the first heartbeat frequency based on the main heartbeat cycle includes: converting the main heartbeat cycle into a second frequency; performing frequency domain analysis on the heartbeat signal waveform to obtain a second spectrum; determining the dominant heartbeat frequency of the heartbeat signal waveform based on the second spectrum; and determining the first heartbeat frequency based on the second frequency and the dominant heartbeat frequency. Performing frequency domain analysis on the signal waveform allows for the extraction of the most energetic dominant frequency through Fourier transform, reflecting the frequency stability of the signal. The fusion of the two analysis methods can compensate for the shortcomings of a single dimension (for example, the time domain is easily affected by sudden noise, and the frequency domain is prone to losing instantaneous features), and obtain more reliable first respiratory rate and first heart rate.

[0096] Frequency domain analysis is performed on the signal waveform to extract energy concentration features from the signal's frequency distribution. First, a Fourier transform is performed on the signal waveform to convert the time-domain signal into a frequency-domain distribution, resulting in a spectrum representing the frequency distribution. Then, within their respective effective frequency ranges (e.g., 0.2-0.34Hz for respiration, 0.8-2.5Hz for heartbeat), the frequency distribution with the strongest energy is identified and determined as the dominant frequency in the frequency domain, reflecting the energy concentration characteristics of the signal in the frequency dimension. Finally, the dominant period is converted into its corresponding frequency and fused with the dominant frequency in the frequency domain. A weighted fusion method can be used to avoid noise interference from a single method, thereby obtaining the first respiratory frequency and the first heartbeat frequency.

[0097] Referring to steps S1033 and S1034, in this embodiment, the respiratory reference frequency and heart rate reference frequency are obtained through statistical calculation of time-domain peak intervals. Taking the respiratory reference frequency as an example, firstly, time-domain peak filtering is performed to obtain the corresponding first effective peak. The constant false alarm rate (CFAR) algorithm can be used to detect effective peaks. Effective peaks are filtered by setting parameters, such as setting the amplitude factor to 1.5 and the minimum group size to 3. This yields the CFAR peak detection amplitude threshold = 1.5 × median, the minimum number of effective peaks (min_group_size) = 3, and the tolerance grouping can be set to 10%. Specifically, the time interval characteristics of the effective peaks can be calculated first, i.e., the time interval between every two effective peaks, such as using intervals to represent the time interval, t... n Let intervals represent the time corresponding to the nth valid peak, then intervals = [t1−t0, t2−t1, ..., t n -t n−1 Then, the time intervals are grouped, allowing for a 10% margin of error; for example, time intervals between 0.9 seconds and 1.1 seconds can be grouped together. The median of the largest group is calculated as interval_median = median(largest_group), and the reference frequency can then be calculated, expressed as the reciprocal of the median. The corresponding calculation process for the respiratory reference frequency is as described above and will not be detailed here.

[0098] For example, the detected peak heart rate time is [1.0, 1.8, 2.1, 3.0, 3.9, 4.0, 5.2] seconds; after CFAR filtering, the effective peak values ​​are [1.0, 2.1, 3.0, 5.2] (points with too low amplitude are removed); the peak interval is obtained as [1.1, 0.9, 2.2] seconds; among them, the median of the largest group (0.9~1.1s) is taken as 1.0 seconds, and the heart rate reference frequency is calculated as 1 / 1.0 = 1.0 Hz (60 bpm).

[0099] After obtaining the first respiratory rate and respiratory reference rate corresponding to the respiratory signal, and the first heart rate and heart rate reference rate corresponding to the heart rate signal, step S1035 can be executed, that is, determining the confidence level based on the deviation between the first respiratory rate and the respiratory reference rate, and the deviation between the first heart rate and the heart rate reference rate. The confidence level is used to characterize the reliability of the first respiratory rate and the first heart rate, which are used to characterize the breathing and heartbeat of the target user.

[0100] The confidence level is determined by analyzing the deviations between the first respiratory rate and the respiratory reference rate, as well as the deviations between the first heart rate and the heart rate reference rate. Small deviations indicate a good match between the two analyses, suggesting high accuracy of the calculated respiratory and heart rates, i.e., high confidence.

[0101] In one embodiment of this application, determining the confidence level based on the deviation between the first respiratory rate and the respiratory reference rate and the deviation between the first heart rate and the heart rate reference rate includes: calculating the first deviation between the first respiratory rate and the respiratory rate reference rate and the second deviation between the first heart rate and the heart rate reference rate; counting the first number of the first effective peak and the second number of the second effective peak; and determining the confidence level based on the first deviation, the second deviation, the first number, and the second number.

[0102] The reference frequency is derived from the effective peak value of the signal waveform (reflecting the objective characteristics of the signal waveform). The smaller the deviation between the first respiratory frequency and the respiratory reference frequency, and the smaller the deviation between the first heartbeat frequency and the heartbeat reference frequency, the more consistent it is with the actual characteristics of the signal, and the higher the reliability. Simultaneously, the number of effective peak values ​​should be counted to clarify the significance of the signal characteristics. A higher number indicates less noise interference and more stable periodicity; for example, five effective peak values ​​reflect the true period better than two. By calculating the deviation between these two values ​​and statistically analyzing the effective peak values, a comprehensive reflection of signal quality can be achieved, avoiding the bias of a single indicator. Quantitative calculation rules ensure comparability and stability of confidence levels, improving the accuracy of reliability assessment.

[0103] After obtaining the confidence level, multi-distance unit filtering in the spatial dimension can be used to solve the problem of signal interference in a single region, ensuring that the output result comes from the region with the best signal quality. In some embodiments of this application, the process of outputting at least a second respiratory frequency and a second heart rate based on the first respiratory frequency, the first heart rate, and the confidence level includes:

[0104] Based on the confidence level of each distance unit corresponding to the echo signal, an effective distance unit is determined; the first respiratory frequency and the first heart rate corresponding to the effective distance unit are respectively determined as the second respiratory frequency and the second heart rate; the second respiratory frequency and the second heart rate are output. Each distance unit corresponds to a specific distance interval between the signal source and the target user.

[0105] Taking radar as an example, different spatial regions within the radar's detection range correspond to different range units. The first respiratory rate, first heart rate, and confidence level are calculated for each range unit. Since the target user's physiological signal resides within their respective range unit, while other areas may experience noise or interference (such as signal reflection from furniture), parallel processing of multiple range units can preserve the optimal signal. A corresponding confidence threshold can be set, and range units with confidence levels higher than this threshold can be selected as candidate range units. Then, the range unit with the highest confidence among the candidate range units can be taken as the effective range unit, and its corresponding respiratory rate and heart rate can be output.

[0106] To output more accurate respiratory and heart rates, embodiments of this application can output corresponding frequencies based on confidence levels. In some embodiments of this application, at least a second respiratory rate and a second heart rate are output based on a first respiratory rate, a first heart rate, and a confidence level, including: if the confidence level is less than a preset threshold, determining the previously output respiratory and heart rates as the second respiratory and heart rates, and outputting the second respiratory and heart rates; if the confidence level is not less than the preset threshold, determining the first respiratory rate as the second respiratory rate, and determining the first heart rate as the second heart rate; and outputting the second respiratory and heart rates.

[0107] Since signal detection is real-time or has a certain detection period (the detection period usually corresponds to a short time), in order to ensure the validity and reliability of data output, when a low confidence level is detected, the frequency information corresponding to the current low confidence level may not be output. Instead, the frequency information of the previous output that is closest to the current high confidence level may be output.

[0108] To allow users to perceive the reliability of the test results, the confidence level can be converted into user-perceptible error information and output synchronously with the detected frequency information. In one implementation, at least a second respiratory rate and a second heart rate are output based on a first respiratory rate, a first heart rate, and a confidence level, including: determining the first respiratory rate as the second respiratory rate and the first heart rate as the second heart rate; converting the confidence level into perceptible error information; and outputting the second respiratory rate, the second heart rate, and the perceptible error information.

[0109] When converting confidence levels into perceptible error information, the format of this perceptible error information can be determined based on the current output mode. For example, when displaying the target user's breathing and heart rate via an electronic device's display component, the perceptible error information can be visual information, such as corresponding text descriptions displayed in a specific font color. For instance, when the confidence level is high, the breathing and heart rate outputs are displayed in the default font; when the confidence level is low, the breathing and heart rate are displayed in gray font. Correspondingly, if it is voice-broadcast frequency information, the perceptible error information can be presented through prompts or voice messages. If confidence level C represents a certain level, a corresponding confidence threshold can be set, and the perceptible error information can be determined based on the relationship between the current confidence level C and the confidence threshold. For example, when C ≥ 0.8, the perceived error information obtained could be: error ±1 time / minute, reliable result, allowing users to know that high confidence corresponds to good consistency and a small error range; if 0.5 ≤ C < 0.8, the corresponding perceived error information could be: error ±2 times / minute, relatively reliable result, allowing users to know that medium confidence results have slight deviations and a moderate error range; if 0.3 ≤ C < 0.5, the corresponding perceived error information is: error ±3 times / minute, for reference only, as low confidence results have large deviations, prompting users to refer to the results with caution; if C < 0.3, the corresponding perceived error information is: poor signal quality and large error, used to prompt users that the results are unreliable and avoid misjudgment.

[0110] The embodiments of this application will be described below with reference to the corresponding waveform diagrams. It should be noted that the waveforms in this embodiment are merely examples for ease of description, and the actual waveform effects need to be determined based on the actual application scenario and the characteristics of the corresponding acquisition and processing equipment. This application does not impose any limitations on this.

[0111] See Figure 3 The illustration shows a schematic diagram of an echo signal provided in an embodiment of this application. Taking a UWB chip as an example, the UWB chip transmits pulses with continuous phase and receives echoes to obtain the original channel impulse response (CIR) signal. This is a two-dimensional signal containing phase and amplitude information at different distances (fast time dimension) and different time points (slow time dimension). This signal contains reflections from all targets and is relatively cluttered. The phase transmitted by the UWB chip is a continuous time point, and the echoes at the same location can reflect the periodic motion at that location. Thus, by determining the frequency components contained in the echo signal, it is possible to reflect what kind of periodic motion exists at that location. Figure 3 The vertical axis represents fast time, which is the time it takes for the pulse signal emitted by the UWB chip to propagate, and its unit is nanoseconds (ns); the horizontal axis represents slow time, which is the radar's detection time of the target, and its unit is seconds (s).

[0112] Based on the Time of Flight (ToF) principle, specific distance units can be filtered out (e.g., the location of a user's chest cavity). Therefore, signal changes of these distance units in the slow time dimension can be extracted, i.e., phase or amplitude changes at the same distance over consecutive time points. For example, by analyzing... Figure 3 (Slow-time-fast-time two-dimensional spectrum) analysis is performed, first identifying the range gate region where signal energy is concentrated (e.g., Figure 3 The yellow-green band in the 0-10 time interval of the medium-fast timeframe can represent the strong signal region corresponding to the human chest cavity reflex. Based on this, a corresponding time segment is extracted along the slow time axis, and multiple sets of distance cells (e.g., ...) are extracted from this energy concentration region. Figure 4 and Figure 5 The time-domain data for Tracks 6-11 shown is required. Since UWB signals are inherently complex signals, the extracted time-domain data needs to be further decomposed into a real part (the physically measurable signal amplitude component) and an imaginary part (the component orthogonal to the real part, describing the signal's phase characteristics). After filtering to remove high-frequency electromagnetic interference, static noise, and other noise, the data is generated. Figure 4 and Figure 5 The multi-channel time-domain waveform is shown. Among them, Figure 4 This paper illustrates a multi-channel time-domain waveform diagram of the real components of a UWB signal provided in an embodiment of this application. Figure 5 This paper illustrates a multi-channel time-domain phase diagram of the imaginary component of a UWB signal provided in an embodiment of this application.

[0113] exist Figure 4 and Figure 5 In the diagram, the horizontal axis represents time, which is the duration of continuous data collection representing the slow time dimension. Figure 4 The vertical axis represents the amplitude corresponding to the real part. Figure 5 The vertical axis represents the phase corresponding to the imaginary component. Tracks 6-11 characterize the waveforms obtained for different distance units. Figure 4 The real part waveform shows the change in signal strength of each channel with slow time. Figure 5 The imaginary waveform reflects the temporal evolution of the signal phase characteristics, providing a foundation for the subsequent extraction of the periodic and phase modulation information of human micro-movements (breathing / heartbeat).

[0114] based on Figure 4 and Figure 5 It can filter out signals containing micro-movement characteristics of the human body, for Figure 4 The real part signal shown and Figure 5The imaginary part of the signal shown can be further processed to extract respiratory and heart rate components. For example, by using bandpass filtering, the 0.2-0.34Hz respiratory signal and the 1-2Hz heart rate signal can be separated. Therefore, the following can be obtained: Figure 6 The diagram shown is a schematic of the filtered respiratory signal waveform. Figure 6 The image shows the filtered respiratory signal waveforms of the distance cells corresponding to Tracks 6-11. Similarly, the following results were obtained: Figure 7 The diagram shows a filtered heartbeat signal waveform. Figure 6 Further data characterizing human respiratory rate can be extracted and labeled as respiratory peaks. Figure 6 The waveform in the image is represented by red dots. Correspondingly, in... Figure 7 Further data representing human heart rate can be extracted and labeled as heartbeat peaks. Figure 7 The waveforms in the image are represented by blue dots.

[0115] Taking respiratory signals as an example, in determining the dominant respiratory frequency (i.e., the first respiratory frequency described in the aforementioned embodiments) within the current time window (e.g., a 20-second time window), different distance units can be processed separately to obtain the dominant respiratory frequency corresponding to each distance unit. Then, effective respiratory peaks are filtered based on the respiratory signal waveform, and a respiratory reference frequency is determined based on the filtered effective respiratory peaks. The deviation between the dominant respiratory frequency and the respiratory reference frequency for each distance unit is calculated, and the confidence level corresponding to that distance unit is determined based on this deviation. The smaller the deviation, the higher the confidence level. After obtaining the confidence levels for each distance unit (e.g., Track 6-Track 11), the unit with the highest confidence level is taken as the optimal distance unit within the current time window, and the dominant respiratory frequency corresponding to that distance unit is used to determine the user's respiratory frequency within that time window.

[0116] Furthermore, curves showing the changes in respiratory and heart rate over time can be output. After selecting the optimal distance unit corresponding to each time window, the frequency data of that distance unit can be used as the data basis for outputting the curves showing the changes in respiratory and heart rate over time, thus obtaining a complete monitoring cycle's curves showing the changes in respiratory and heart rate over time. For example, the respiratory rate corresponding to the optimal distance unit determined in the time window of 0-20 seconds (i.e., the distance unit with the highest confidence in that time window) can be used as the data basis for the 20-second time point in the monitoring cycle of the curve to be output, and the respiratory rate of the optimal distance unit determined in the time window of 5-25 seconds can be used as the data basis for the 25-second time point in the curve to be output. Specific details are explained in subsequent embodiments and will not be elaborated here.

[0117] In this way, an optimal distance unit can be calculated for each time window, and the respiratory rate calculated by the distance unit is used as the respiratory rate to be output for the corresponding time window, so as to obtain the user's respiratory rate curve within the monitoring period (including multiple time windows).

[0118] The following is combined with Figures 6-8 The process of outputting waveforms representing the user's respiratory and heart rates is explained. Respiratory rate is used primarily as an example; the processing of heart rate is similar and will not be detailed further. See [link to documentation]. Figure 6 ,exist Figure 6 Red dots representing respiratory peaks were selected from the data, and then each distance unit (e.g.) was analyzed. Figure 6 The time intervals (also known as peak widths) between respiratory peaks shown in Tracks 6-11 are statistically analyzed. A certain tolerance parameter can be set. For example, when the tolerance parameter is 0.1, peak widths within 0% can be considered as the same group. The time interval between the most common respiratory peaks is determined as the main respiratory cycle. Then, the main respiratory frequency A corresponding to this main cycle is calculated. The time interval between the most common respiratory peaks can refer to the time interval of effective groups that exceed a certain number threshold. One group corresponds to two respiratory peaks. For example, there are 6 respiratory peak groups with a time interval of 5, which is the most numerous and exceeds the number threshold of 3. Therefore, 5 is determined as the main cycle.

[0119] Then calculate the respiratory reference frequency B, such as Figure 6 The diagram shows the identification of each respiratory peak ( Figure 6 After the red dot in the middle, for each distance unit (such as...) Figure 6 The amplitudes of the respiratory peaks shown in Tracks 6-11 are calculated, and then the median of all amplitudes is taken. A floating parameter can be set, and the product of the floating parameter and the median is calculated. This product is used as the amplitude threshold. Respiratory peaks with amplitudes less than the amplitude threshold are filtered out, and the remaining respiratory peaks are taken as valid respiratory peaks. Then, the interval width of the valid respiratory peaks is calculated as the reference cycle of breathing, and the respiratory reference frequency B is obtained based on the reference cycle.

[0120] Then, the deviation between the dominant respiratory frequency A and the respiratory reference frequency B for each distance unit is calculated. The confidence level is determined based on this deviation; the smaller the deviation, the higher the confidence level. For example, the confidence level for distance unit Track 6 is C6, for Track 7 it is C7, and so on, until the confidence level for Track 11 is calculated to be C11. It should be noted that the confidence levels are calculated for signal data within the same time window. The confidence levels of all distance units are compared, and the distance unit with the highest confidence level is selected as the optimal distance unit for the current time window. The dominant respiratory frequency calculated from this optimal distance unit is determined as the user's respiratory frequency to be output. The processing of heart rate is similar to that of respiratory rate and will not be detailed here.

[0121] In this embodiment, the length of the time window can be at least one complete physiological signal cycle to avoid incomplete cycle statistics, while the duration should not be too long to ensure real-time performance. The time window can be a sliding window (non-fixed interval window), where the step size of the sliding window is smaller than the window duration, such as a 5-second window with a step size of 1 second. The time windows can also overlap to ensure the correlation between data from adjacent windows.

[0122] For example, in this embodiment, the window duration of each time window can be 20 seconds, and the sliding step size is 5 seconds. This ensures sufficient data for frequency analysis while guaranteeing a data update every 5 seconds, resulting in a higher output frequency, more continuous curves, and faster reflection of changes in vital signs. See also... Figure 8 In the graph of respiratory rate versus time, the raw data corresponding to the 20-second mark on the time axis comes from the respiratory rate data of the optimal distance unit (such as Track 9) selected within a time window of 0-20 seconds. Figure 8 The raw data corresponding to the 25-second mark on the mid-time axis can be respiratory rate data from the optimal distance unit (such as Track 10) selected within a time window of 5-25 seconds, and so on, to obtain... Figure 8 The original data is used. In this process, the time intervals on the time axis in the graph of respiratory rate changing over time can be set according to the step size of the time window. In this way, each point in time represents the analysis result of a complete time window, which can achieve the purpose of real-time and high-precision monitoring of vital signs. Similarly, the graph of heart rate changing over time can be obtained.

[0123] exist Figure 8In this context, the line represented by the original data is obtained after calculating frequency A, frequency B, and confidence level, and then filtering for optimal distance units based on the confidence level. However, these are also susceptible to short-term noise and random errors. Therefore, to reflect the true trend of frequency fluctuations, in this embodiment, a corresponding smooth curve can also be generated based on the original data, such as... Figure 8 As shown, this reduces the error range and fluctuations in the original data, making the data more continuous, more in line with the laws of physiological signals, and more meaningful for visualization.

[0124] join Figure 9 This illustration shows a signal processing diagram of an application scenario provided by an embodiment of this application. For example, in a home environment, to monitor the breathing and heartbeat of an adult during sleep, a UWB (Ultra Wideband) radar device is used for non-contact detection. The device is installed 1.5 meters above the head of the bed and can cover multiple distance gates (distance units) where the human body is located, outputting physiological parameters (such as respiratory rate and heart rate) in real time.

[0125] Figure 9 The original signal shown is the echo signal of the received pulse signal. The device transmits one UWB radar frame at a time, and each frame contains multiple radar pulses. The UWB chip receives signals simultaneously with transmission, with a sampling frequency of 1 GHz, meaning the interval between two receptions (fast time) is 1 ns. According to the Time of Flight (ToF) formula d = c*t / 2 (where c is the speed of light), the distance corresponding to the time interval between two receptions is 0.15 m, meaning the interval between each range gate is 0.15 m. The interval between two UWB radar frames is called slow time.

[0126] Data from multiple range gates surrounding the measured distance are simultaneously processed using a sliding window, with calculations performed every 5 seconds using the data from the previous 20 seconds to ensure data timeliness and stability. For each range gate, the phase of the CIR (Channel Impulse Response) signal is extracted and calculated during slow time. Then... Figure 9 As shown, the original signal is filtered and preprocessed, specifically, a second-order high-pass or low-pass filter is used to remove baseline drift less than 0.05 Hz and noise greater than 5 Hz, resulting in a preprocessed signal. Then, respiratory and heartbeat signals are separated and extracted from the preprocessed signal.

[0127] In the extraction of the corresponding respiratory signal, the preprocessed signal is first subjected to a second-order bandpass filter (lowcut=0.2Hz, highcut=0.34Hz, corresponding to 12~20 breaths per minute) to obtain the respiratory component (filtered_respiratory). In the extraction of the heartbeat signal, the respiratory component is subtracted from the preprocessed signal (preprocessed_signal - filtered_respiratory) to obtain the basic heartbeat signal (heartbeat_base), which is then subjected to a fourth-order bandpass filter (lowcut=0.8Hz, highcut=2.5Hz, corresponding to 48~150 heartbeats per minute) to obtain the heartbeat component (filtered_heartbeat).

[0128] The most common interval width between peak values ​​is calculated for both respiratory and heart rate signals to obtain the dominant frequency A. A grouping tolerance of 0.1 is set, meaning peak widths with an error within 10% are considered to belong to the same group. A group size of 3 (min_group_size) means only groups containing more than 3 intervals are considered valid. For example, if multiple peak values ​​are detected in the respiratory component, after calculating the intervals between adjacent peaks, they are grouped with a tolerance of 0.1. The group with the most intervals is selected as the valid group, and the frequency corresponding to its average interval is the respiratory dominant frequency A. The corresponding heart rate dominant frequency A is calculated in the same way.

[0129] The CFAR (Constant False Alarm Rate) method was used on both respiratory and heart rate signals to find the selected peaks, and the reference frequency B was calculated. First, the amplitude of each peak and the median amplitude of all peaks were calculated. The amplitude factor was multiplied by the median to obtain the amplitude threshold. Peaks smaller than the amplitude threshold were filtered out, resulting in the selected peaks. The frequency B was calculated based on the intervals between these peaks. Here, the amplitude is the peak amplitude, calculated as the difference between the peak and its adjacent trough.

[0130] The distance gate with the closest frequencies A and B among all distance gates is selected, thus obtaining the distance gate with the highest confidence. Frequency A of this distance gate is then used as the final output respiratory or heart rate. This device can accurately monitor respiratory and heart rates during sleep without contact with the human body, and it does not require pre-determining the distance gate where the human body is located. It can adapt to scenarios where the human body is moving, providing a convenient and reliable solution for home health monitoring.

[0131] See Figure 10 This application also provides a signal processing system, which includes a signal device 401 and a processing device 402.

[0132] The signal device 401 is used to continuously transmit pulse signals, which are used to detect the breathing and heartbeat of the target user; and to receive the echo signal of the pulse signals.

[0133] The processing device 402 is configured to process the echo signal to obtain a first respiratory frequency characterizing the breathing of the target user and a first heart rate characterizing the heartbeat of the target user, and a confidence level characterizing the first respiratory frequency and the first heart rate as a reliability of the breathing and heartbeat of the target user; and to output at least a second respiratory frequency and a second heart rate based on the first respiratory frequency, the first heart rate and the confidence level.

[0134] In some embodiments, obtaining a first respiratory rate characterizing the breathing of the target user and a first heart rate characterizing the heartbeat of the target user based on the echo signal processing, and a confidence level characterizing the first respiratory rate and the first heart rate as a representation of the reliability of the breathing and heartbeat of the target user, includes:

[0135] Based on the echo signal, the respiratory signal waveform and the heartbeat signal waveform are obtained respectively;

[0136] Time-domain analysis was performed on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first respiratory frequency and the first heartbeat frequency;

[0137] The time-domain peak values ​​of the respiratory signal waveform and the heartbeat signal waveform are filtered respectively to obtain the first effective peak value corresponding to the respiratory signal waveform and the second effective peak value corresponding to the heartbeat signal waveform;

[0138] The respiratory reference frequency and the heart rate reference frequency are determined based on the first time interval characteristics of the first effective peak value and the second time interval characteristics of the second effective peak value, respectively.

[0139] A confidence level is determined based on the deviation between the first respiratory rate and the respiratory reference rate, and the deviation between the first heart rate and the heart rate reference rate. The confidence level is used to characterize the reliability of the first respiratory rate and the first heart rate as characterizes the breathing and heartbeat of the target user.

[0140] In some embodiments, obtaining the respiratory signal waveform and the heartbeat signal waveform based on the echo signal includes:

[0141] Based on the echo signal, a two-dimensional signal array containing a distance dimension and a time series dimension is generated; the distance dimension is divided into multiple distance units based on a fixed distance interval, and each distance unit corresponds to a specific distance interval between the signal source and the target user; the time series dimension consists of continuous time frames, and each time frame corresponds to a complete pulse transmission and reception cycle.

[0142] For each distance unit, the phase change sequence of the distance unit in the time series dimension is extracted as the first feature waveform;

[0143] The first characteristic waveform is processed by a filter corresponding to the first frequency of the breathing characteristic to obtain the breathing signal waveform;

[0144] For each time frame, the amplitude distribution sequence of the time frame in the distance dimension is extracted as the second feature waveform;

[0145] The second characteristic waveform is processed by a filter corresponding to the second frequency of the heartbeat characteristic to obtain the heartbeat signal waveform.

[0146] In some embodiments, performing time-domain analysis on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first respiratory frequency and the first heartbeat frequency includes:

[0147] Perform time-domain analysis on the respiratory signal waveform to obtain all peak points in the respiratory signal waveform, and calculate the first time interval between adjacent peak points;

[0148] The main respiratory cycle is determined based on the frequency of occurrence of the first time interval;

[0149] The dominant frequency of the respiratory frequency domain is determined based on the frequency components in the first spectrum diagram;

[0150] Based on the main respiratory cycle, a first respiratory rate is determined;

[0151] Perform time-domain analysis on the heartbeat signal waveform to obtain all peak points in the heartbeat signal waveform, and calculate the second time interval between adjacent peak points;

[0152] The main heartbeat cycle is determined based on the frequency of occurrence of the second time interval;

[0153] The first heartbeat frequency is determined based on the main heartbeat cycle.

[0154] In some embodiments, determining the first respiratory rate based on the master respiratory cycle includes:

[0155] Convert the main respiratory cycle into a first frequency;

[0156] Frequency domain analysis was performed on the respiratory signal waveform to obtain the first spectrum.

[0157] Based on the first spectrum diagram, determine the dominant frequency of the respiratory signal waveform in the respiratory frequency domain;

[0158] The first respiratory frequency is determined based on the first frequency and the dominant frequency of the respiratory frequency domain.

[0159] In some embodiments, determining the confidence level based on the deviation between the first respiratory rate and the respiratory reference rate, and the deviation between the first heart rate and the heart rate reference rate, includes:

[0160] Calculate the first deviation between the first respiratory rate and the respiratory rate reference rate, and the second deviation between the first heart rate and the heart rate reference rate, respectively;

[0161] Count the first number of the first effective peak and the second number of the second effective peak;

[0162] The confidence level is determined based on the first deviation, the second deviation, the first quantity, and the second quantity.

[0163] In some embodiments, the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level includes:

[0164] Based on the confidence level of each distance unit corresponding to the echo signal, an effective distance unit is determined; wherein, each distance unit corresponds to a specific distance interval between the signal source and the target user;

[0165] The first respiratory rate and the first heart rate corresponding to the effective distance unit are respectively determined as the second respiratory rate and the second heart rate;

[0166] Output the second respiratory rate and the second heart rate.

[0167] In some embodiments, the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level includes:

[0168] If the confidence level is less than a preset threshold, the previously output respiratory rate and heart rate are determined as the second respiratory rate and the second heart rate, and the second respiratory rate and the second heart rate are output.

[0169] If the confidence level is not less than the preset threshold, the first respiratory rate is determined as the second respiratory rate, and the first heart rate is determined as the second heart rate; the second respiratory rate and the second heart rate are output.

[0170] In some embodiments, the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level includes:

[0171] The first respiratory rate is determined as the second respiratory rate, and the first heart rate is determined as the second heart rate;

[0172] The confidence level is converted into perceptible error information;

[0173] Output the second respiratory rate, the second heart rate, and the perceptible error information.

[0174] It should be noted that the specific implementation of the signal device and processing device in this embodiment can be referred to the corresponding content above, and will not be described in detail here.

[0175] In another embodiment of this application, a readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the signal processing method as described above.

[0176] In another embodiment of this application, an electronic device is also provided, which may include:

[0177] Memory, used to store applications and the data generated by the running of the applications;

[0178] A processor for executing the application to implement the signal processing method described above.

[0179] It should be noted that the specific implementation of the processor in this embodiment can be referred to the corresponding content above, and will not be described in detail here.

[0180] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0181] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0182] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0183] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A signal processing method, comprising: Continuously transmit pulse signals, which are used to detect the target user's breathing and the target user's heartbeat; Receive the echo signal of the pulse signal; Based on the echo signal processing, a first respiratory rate for characterizing the breathing of the target user and a first heart rate for characterizing the heartbeat of the target user are obtained, and a confidence level for characterizing the first respiratory rate and the first heart rate is used to characterize the reliability of the breathing and heartbeat of the target user. Based on the first respiratory rate, the first heart rate, and the confidence level, at least the second respiratory rate and the second heart rate are output.

2. The method according to claim 1, wherein obtaining a first respiratory rate for characterizing the breathing of the target user and a first heart rate for characterizing the heartbeat of the target user based on the echo signal processing, and a confidence level for characterizing the first respiratory rate and the first heart rate as a representation of the reliability of the breathing and heartbeat of the target user, comprises: Based on the echo signal, the respiratory signal waveform and the heartbeat signal waveform are obtained respectively; Time-domain analysis was performed on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first respiratory frequency and the first heartbeat frequency; The time-domain peak values ​​of the respiratory signal waveform and the heartbeat signal waveform are filtered respectively to obtain the first effective peak value corresponding to the respiratory signal waveform and the second effective peak value corresponding to the heartbeat signal waveform; The respiratory reference frequency and the heart rate reference frequency are determined based on the first time interval characteristics of the first effective peak value and the second time interval characteristics of the second effective peak value, respectively. A confidence level is determined based on the deviation between the first respiratory rate and the respiratory reference rate, and the deviation between the first heart rate and the heart rate reference rate. The confidence level is used to characterize the reliability of the first respiratory rate and the first heart rate as characterizes the breathing and heartbeat of the target user.

3. The method according to claim 2, wherein obtaining the respiratory signal waveform and the heartbeat signal waveform based on the echo signal comprises: Based on the echo signal, a two-dimensional signal array containing a distance dimension and a time series dimension is generated; the distance dimension is divided into multiple distance units based on a fixed distance interval, and each distance unit corresponds to a specific distance interval between the signal source and the target user; the time series dimension consists of continuous time frames, and each time frame corresponds to a complete pulse transmission and reception cycle. For each distance unit, the phase change sequence of the distance unit in the time series dimension is extracted as the first feature waveform; The first characteristic waveform is processed by a filter corresponding to the first frequency of the breathing characteristic to obtain the breathing signal waveform; For each time frame, the amplitude distribution sequence of the time frame in the distance dimension is extracted as the second feature waveform; The second characteristic waveform is processed by a filter corresponding to the second frequency of the heartbeat characteristic to obtain the heartbeat signal waveform.

4. The method according to claim 2, wherein performing time-domain analysis on the respiratory signal waveform and the heartbeat signal waveform respectively to obtain the first respiratory frequency and the first heartbeat frequency comprises: Perform time-domain analysis on the respiratory signal waveform to obtain all peak points in the respiratory signal waveform, and calculate the first time interval between adjacent peak points; The main respiratory cycle is determined based on the frequency of occurrence of the first time interval; The dominant frequency of the respiratory frequency domain is determined based on the frequency components in the first spectrum diagram; Based on the main respiratory cycle, a first respiratory rate is determined; Perform time-domain analysis on the heartbeat signal waveform to obtain all peak points in the heartbeat signal waveform, and calculate the second time interval between adjacent peak points; The main heartbeat cycle is determined based on the frequency of occurrence of the second time interval; The first heartbeat frequency is determined based on the main heartbeat cycle.

5. The method according to claim 4, wherein determining the first respiratory rate based on the main respiratory cycle comprises: Convert the main respiratory cycle into a first frequency; Frequency domain analysis was performed on the respiratory signal waveform to obtain the first spectrum. Based on the first spectrum diagram, determine the dominant frequency of the respiratory signal waveform in the respiratory frequency domain; The first respiratory frequency is determined based on the first frequency and the dominant frequency of the respiratory frequency domain.

6. The method according to claim 4, wherein determining the confidence level based on the deviation between the first respiratory rate and the respiratory reference rate and the deviation between the first heart rate and the heart rate reference rate comprises: Calculate the first deviation between the first respiratory rate and the respiratory rate reference rate, and the second deviation between the first heart rate and the heart rate reference rate, respectively; Count the first number of the first effective peak and the second number of the second effective peak; The confidence level is determined based on the first deviation, the second deviation, the first quantity, and the second quantity.

7. The method according to claim 1, wherein the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level comprises: Based on the confidence level of each distance unit corresponding to the echo signal, an effective distance unit is determined; wherein, each distance unit corresponds to a specific distance interval between the signal source and the target user; The first respiratory rate and the first heart rate corresponding to the effective distance unit are respectively determined as the second respiratory rate and the second heart rate; Output the second respiratory rate and the second heart rate.

8. The method according to claim 1, wherein the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level comprises: If the confidence level is less than a preset threshold, the previously output respiratory rate and heart rate are determined as the second respiratory rate and the second heart rate, and the second respiratory rate and the second heart rate are output. If the confidence level is not less than the preset threshold, the first respiratory rate is determined as the second respiratory rate, and the first heart rate is determined as the second heart rate. Output the second respiratory rate and the second heart rate.

9. The method according to claim 8, wherein the step of outputting at least a second respiratory rate and a second heart rate based on the first respiratory rate, the first heart rate, and the confidence level comprises: The first respiratory rate is determined as the second respiratory rate, and the first heart rate is determined as the second heart rate; The confidence level is converted into perceptible error information; Output the second respiratory rate, the second heart rate, and the perceptible error information.

10. A signal processing system, comprising: Signaling devices and processing devices; The signaling device is used to continuously transmit pulse signals, which are used to detect the target user's breathing and the target user's heartbeat. Receive the echo signal of the pulse signal; The processing device is configured to obtain, based on the echo signal processing, a first respiratory frequency for characterizing the breathing of the target user and a first heart rate for characterizing the heartbeat of the target user, and a confidence level for characterizing the first respiratory frequency and the first heart rate as a representation of the reliability of the breathing and heartbeat of the target user. Based on the first respiratory rate, the first heart rate, and the confidence level, at least the second respiratory rate and the second heart rate are output.

Images