Intelligent home anti-interference human existence detection method and system based on 24ghz radar

Abstract

The application relates to the technical field of smart homes, and discloses a smart home anti-interference human body existence detection method and system based on a 24GHz radar, which comprises the following steps: collecting intermediate frequency signals in a front space of a smart home based on a 24GHz FMCW radar front-end module integrated in the smart home, and performing filtering processing on the intermediate frequency signals; extracting a distance-velocity matrix of the filtered signals in a motion detection channel, so as to generate a motion mark in the front space through the motion detection channel and the distance-velocity matrix; outputting breathing harmonic energy in the front space through a vital sign channel; sampling a noise spectrum baseline in the front space, and converting the noise spectrum baseline into a noise interference mark through a frequency domain mutual exclusion verifier; and detecting whether a human body exists in the front space according to the motion mark, the breathing harmonic energy and the noise interference mark, so as to obtain a human body existence detection result. The application can achieve both static human body detection reliability and anti-interference performance.

Description

Technical Field

[0001] This invention relates to a smart home anti-interference human presence detection method and system based on 24GHz radar, belonging to the field of smart home technology. Background Technology

[0002] 24GHz frequency modulated continuous wave (FMCW) radar is gradually becoming a core sensor for human presence detection in smart homes due to its advantages such as strong penetration, high privacy and security, and low power consumption. It transmits linear frequency modulated waves and receives the reflected signals from the target to analyze the target's distance, speed, and micro-motion characteristics, providing contactless sensing capabilities for scenarios such as lighting and security.

[0003] Current radar detection schemes mainly rely on Doppler motion characteristics. However, traditional motion detection channels can only identify targets with speeds greater than 0.2 m / s. When a human body is stationary, the micro-motion signals from the rise and fall of the chest cavity are easily drowned out by noise due to their weak energy, causing the determination of the presence of a static human body to fail. The low-frequency harmonics generated by mechanical vibrations such as air conditioners and fans overlap with the human breathing frequency band. Existing schemes lack the ability to specifically identify noise frequency bands, causing mechanical vibrations to be misjudged as the presence of a human body.

[0004] Therefore, the separation of motion detection channels and vital sign channels, as well as the lack of a static baseline for environmental noise, means that the reliability and anti-interference capabilities of static human body detection cannot be achieved simultaneously. Summary of the Invention

[0005] This invention provides a smart home anti-interference human presence detection method and system based on 24GHz radar, the main purpose of which is to achieve both the reliability and anti-interference of static human detection.

[0006] To achieve the above objectives, the present invention provides a smart home anti-interference human presence detection method based on 24GHz radar, comprising:

[0007] Based on the 24GHz FMCW radar front-end module integrated inside the smart home, the intermediate frequency signal in the space in front of the smart home is collected, and the intermediate frequency signal is filtered to obtain a filtered signal.

[0008] The distance-velocity matrix of the filtered signal is extracted in the motion detection channel, and the motion marker in the forward space is generated by the motion detection channel and the distance-velocity matrix.

[0009] The distance-velocity matrix is ​​input into the vital signs channel to output the respiratory harmonic energy in the space in front through the vital signs channel. The vital signs channel includes a static target window selector, a phase difference decomposer, and a harmonic energy extractor.

[0010] The noise spectrum baseline in the foreground space is sampled, and the noise spectrum baseline is converted into a noise interference flag by a frequency domain mutual exclusion verifier.

[0011] Based on the motion marker, the respiratory harmonic energy, and the noise interference marker, the presence of a human body in the space in front is detected, and the presence of a human body is obtained.

[0012] Optionally, the 24GHz FMCW radar front-end module integrated within the smart home collects intermediate frequency signals in the space in front of the smart home, including:

[0013] The voltage-controlled oscillator, power amplifier, transmitting antenna, receiving antenna, and mixer in the 24GHz FMCW radar front-end module are obtained, wherein the mixer includes a first input terminal, a second input terminal, and an intermediate frequency output terminal;

[0014] A 24 GHz frequency-modulated continuous wave with a linear frequency variation over time is generated by the voltage-controlled oscillator.

[0015] The 24 GHz frequency-modulated continuous wave is amplified by the power amplifier to obtain an amplified continuous wave;

[0016] The amplified continuous wave is radiated into the space in front via the transmitting antenna;

[0017] The received antenna captures the reflected echo signal of the amplified continuous wave in the space in front;

[0018] The reflected echo signal in the receiving antenna is received using the first input terminal;

[0019] The second input terminal is used to receive the 24 GHz frequency-modulated continuous wave from the voltage-controlled oscillator;

[0020] Based on the input-output coupling relationship between the first input terminal, the second input terminal and the intermediate frequency output terminal respectively, the reflected echo signal is mixed with the 24 GHz frequency-modulated continuous wave to form an intermediate frequency signal using the intermediate frequency output terminal.

[0021] Optionally, the step of filtering the intermediate frequency signal to obtain a filtered signal includes:

[0022] The bandwidth and center frequency of the bandpass filter are set based on the upper and lower limits of the intermediate frequency signal;

[0023] After setting the bandwidth and the center frequency, the intermediate frequency signal is input into the bandpass filter to output a filtered signal through the bandpass filter.

[0024] Optionally, extracting the distance-velocity matrix of the filtered signal in the motion detection channel includes:

[0025] The filtered signal is input into the FFT processor of the motion detection channel;

[0026] In the FFT processor, the filtered signal is processed by one-dimensional FFT within a single frequency modulation cycle to obtain the distance-dimensional spectrum within the single frequency modulation cycle;

[0027] The filtered signal is processed by slow-time FFT within the same distance cell over multiple consecutive frequency modulation cycles to obtain the velocity dimension spectrum within the same distance cell.

[0028] Arrange the distance-dimensional spectrum and the velocity-dimensional spectrum into a distance-velocity matrix.

[0029] Optionally, generating the motion marker in the forward space through the motion detection channel and the distance-velocity matrix includes:

[0030] The hardware comparator in the motion detection channel determines whether there is a velocity value in the distance-velocity matrix that is greater than a preset velocity threshold.

[0031] When there is a velocity value greater than a preset velocity threshold in the distance-velocity matrix, a high level is used as a motion marker on the distance cell corresponding to the velocity value greater than the preset velocity threshold in the forward space;

[0032] The low level is used as the motion indicator on the distance cell in the space in front of the velocity value that is not greater than a preset velocity threshold in the distance-velocity matrix.

[0033] Optionally, the step of outputting respiratory harmonic energy in the forward space through the vital signs channel includes:

[0034] Acquire the static target window selector, phase difference decomposer, and harmonic energy extractor in the vital signs channel;

[0035] Use a static target window gating tool to filter target range cells with zero velocity values ​​in the range-velocity matrix;

[0036] The target range cell is decomposed into in-phase and quadrature components using the phase difference decomposer.

[0037] The instantaneous phases corresponding to the in-phase component and the quadrature component are calculated using the following formula:

[0038]

[0039] in, Indicates instantaneous phase, Indicates in-phase components, Indicates orthogonal components;

[0040] The instantaneous phase is differentially processed to obtain the phase difference;

[0041] In the harmonic energy extractor, the phase difference is subjected to bandpass filtering within the target frequency band to obtain breathing harmonic energy.

[0042] Optionally, sampling the noise spectrum baseline in the forward space includes:

[0043] When there is no one in the space in front, the intermediate frequency signal of the unmanned person in the space in front is collected using the 24GHz FMCW radar front-end module.

[0044] Spectral analysis is performed on the unmanned intermediate frequency signal to generate a noise spectrum baseline in the space in front.

[0045] Optionally, converting the noise spectrum baseline into a noise interference flag using a frequency domain mutual exclusion verifier includes:

[0046] In the frequency domain mutual exclusion verifier, the energy percentage of the noise spectrum baseline within the target frequency band is calculated;

[0047] When the energy percentage is not less than the preset percentage, a high level is output to generate a noise interference flag;

[0048] When the energy percentage is less than a preset percentage, a low level is output to generate a noise interference flag.

[0049] Optionally, the step of detecting whether a human body exists in the space in front based on the motion marker, the respiratory harmonic energy, and the noise interference marker, and obtaining a human body presence detection result, includes:

[0050] When the motion indicator is high, the presence of a dynamic human body in the space in front is taken as the detection result of the presence of a human body;

[0051] When the motion indicator is low, if the breathing harmonic energy is greater than the hardware-fixed value and the noise interference indicator is low, the presence of a static human body in the space in front is taken as the detection result of the presence of a human body.

[0052] When the motion indicator is low, if the breathing harmonic energy is not greater than the hardware-fixed value, the absence of a human body in the space in front is taken as the detection result of the presence of a human body.

[0053] When the motion indicator is low, if the noise interference indicator is not low, the absence of a human body in the space in front is taken as the detection result of the presence of a human body.

[0054] To address the aforementioned problems, the present invention also provides a smart home anti-interference human presence detection system based on 24GHz radar, the system comprising:

[0055] The signal filtering module is used to collect intermediate frequency signals in the space in front of the smart home based on the 24GHz FMCW radar front-end module integrated inside the smart home, and to filter the intermediate frequency signals to obtain a filtered signal.

[0056] The marker generation module is used to extract the distance-velocity matrix of the filtered signal in the motion detection channel, so as to generate the motion marker in the front space through the motion detection channel and the distance-velocity matrix;

[0057] An energy output module is used to input the distance-velocity matrix into the vital signs channel to output the respiratory harmonic energy in the space in front through the vital signs channel. The vital signs channel includes a static target window selector, a phase difference decomposer, and a harmonic energy extractor.

[0058] The noise conversion module is used to sample the noise spectrum baseline in the forward space and convert the noise spectrum baseline into a noise interference flag through a frequency domain mutual exclusion verifier.

[0059] The human body detection module is used to detect whether there is a human body in the space in front of the object based on the motion marker, the respiratory harmonic energy and the noise interference marker, and to obtain the detection result of the presence of a human body.

[0060] Compared to the problems described in the background technology, this embodiment of the invention uses a 24GHz FMCW radar front-end module integrated within a smart home to collect intermediate frequency signals in the space in front of the smart home. The intermediate frequency signals are then filtered to obtain a filtered signal, thus solving the problem of high-frequency interference in the original signal causing distortion in feature extraction. Simultaneously, bandpass filtering suppresses out-of-band noise, improves the signal-to-noise ratio, and provides a clean signal source for dual-channel processing. This embodiment extracts the range-velocity matrix of the filtered signal in the motion detection channel, and uses the motion detection channel and the range-velocity matrix to generate motion markers in the space in front, enabling rapid response to moving targets while retaining the real-time advantages of traditional solutions and reducing the false negative rate in dynamic scenes. This embodiment outputs the call signs in the space in front through the vital signs channel. This invention addresses the limitation of static human body micro-motion signals being unable to be separated from environmental vibrations by absorbing harmonic energy. A phase differential decomposer precisely extracts 0.1-0.5Hz vital harmonic energy, overcoming the bottleneck of static human body detection. In this embodiment, the noise spectrum baseline in the foreground space is sampled, and a frequency domain mutual exclusion verifier converts this baseline into a noise interference flag. This resolves the misjudgment caused by the overlap of mechanical vibration harmonics and respiratory signal frequency bands. Dynamic calibration of the noise baseline and frequency band energy ratio thresholds achieve frequency domain decoupling of vibration interference and vital signs. Furthermore, this invention addresses the shortcomings of fragmented motion and static detection logic and poor environmental adaptability by using the motion flag, respiratory harmonic energy, and noise interference flag. The motion flag prioritizes triggering and mutually excludes static vital signs verification, comprehensively improving detection reliability. Therefore, this invention achieves both high reliability and strong anti-interference capabilities in static human body detection. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating a smart home anti-interference human presence detection method based on 24GHz radar according to an embodiment of the present invention.

[0062] Figure 2 A noise spectrum comparison diagram of a smart home anti-interference human presence detection method based on 24GHz radar provided in an embodiment of the present invention;

[0063] Figure 3 This is a system architecture diagram of a smart home anti-interference human presence detection method based on 24GHz radar provided in an embodiment of the present invention.

[0064] Figure 4 This is a schematic diagram of a module for implementing the smart home anti-interference human presence detection system based on 24GHz radar, according to an embodiment of the present invention.

[0065] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0066] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0067] This application provides a smart home anti-interference human presence detection method based on 24GHz radar. The executing entity of the smart home anti-interference human presence detection method based on 24GHz radar includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application embodiment: a server, a terminal, etc. In other words, the smart home anti-interference human presence detection method based on 24GHz radar can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0068] Reference Figure 1 The diagram shown is a flowchart illustrating a smart home anti-interference human presence detection method based on 24GHz radar according to an embodiment of the present invention. In this embodiment, the smart home anti-interference human presence detection method based on 24GHz radar includes:

[0069] S1. Based on the 24GHz FMCW radar front-end module integrated within the smart home, the intermediate frequency signal in the space in front of the smart home is collected, and the intermediate frequency signal is filtered to obtain a filtered signal.

[0070] This invention utilizes a 24GHz FMCW radar front-end module integrated within a smart home system to collect intermediate frequency (IF) signals from the space in front of the smart home. The IF signals are then filtered to obtain a filtered signal, thus addressing the issue of high-frequency interference in the original signal causing distortion in feature extraction. Simultaneously, bandpass filtering suppresses out-of-band noise, improves the signal-to-noise ratio, and provides a clean signal source for dual-channel processing.

[0071] In one embodiment of the present invention, the 24GHz FMCW radar front-end module integrated within a smart home system collects intermediate frequency (IF) signals from the space in front of the smart home system, comprising: acquiring a voltage-controlled oscillator (VCO), a power amplifier, a transmitting antenna, a receiving antenna, and a mixer from the 24GHz FMCW radar front-end module, wherein the mixer includes a first input terminal, a second input terminal, and an IF output terminal; generating a 24GHz frequency-modulated continuous wave with a linearly changing frequency over time using the VCO; amplifying the 24GHz frequency-modulated continuous wave using the power amplifier to obtain an amplified continuous wave; radiating the amplified continuous wave into the space in front of the smart home system via the transmitting antenna; capturing the reflected echo signal of the amplified continuous wave in the space in front of the smart home system using the receiving antenna; receiving the reflected echo signal from the receiving antenna using the first input terminal; receiving the 24GHz frequency-modulated continuous wave from the VCO using the second input terminal; and, based on the input-output coupling relationship between the first input terminal, the second input terminal, and the IF output terminal, using the IF output terminal to couple the reflected echo signal with the 24GHz frequency-modulated continuous wave from the transmitting antenna. GHz frequency modulated continuous wave mixer is an intermediate frequency signal.

[0072] The voltage-controlled oscillator (VCO) is a voltage-controlled oscillator that uses the input voltage to adjust the output frequency, making the output signal frequency change linearly with time, thereby generating a 24 GHz frequency-modulated continuous wave. The power amplifier (PA) is an RF power amplifier used to amplify the 24 GHz frequency-modulated continuous wave output from the VCO using current / voltage amplification to achieve the required transmission power, forming an amplified continuous wave. The transmitting antenna is a microstrip or metal patch antenna, with its electrical port connected to the PA output, radiating the amplified continuous wave as electromagnetic waves into the space in front. The receiving antenna is a microstrip or metal patch antenna with the same structure as the transmitting antenna or designed in pairs, used to receive electromagnetic waves reflected from objects in the space in front, outputting the reflected echo signal. The mixer is a three-port RF device; the first input terminal is connected to the receiving antenna to receive the reflected echo signal, and the second input terminal is connected to the VCO to receive the 24 GHz frequency-modulated continuous wave. The 24 GHz frequency-modulated continuous wave is used as the local oscillator signal. The intermediate frequency output terminal multiplies the two input signals and filters out high-frequency components, directly outputting an intermediate frequency signal with a frequency equal to the instantaneous frequency difference between the two. The input-output coupling relationship refers to the interconnection and communication between the input terminal and the output terminal in the hardware structure. The 24 GHz frequency-modulated continuous wave refers to a continuous sine wave generated by a voltage-controlled oscillator with a frequency that changes linearly with time. The carrier center frequency is 24 GHz, which is used for radar ranging and speed measurement. Its instantaneous frequency changes continuously from low to high within the sweep bandwidth, forming a linear sweep diagonal line.

[0073] In one embodiment of the present invention, the step of filtering the intermediate frequency signal to obtain a filtered signal includes: setting the bandwidth and center frequency of a bandpass filter based on the upper and lower limits of the intermediate frequency signal; and after setting the bandwidth and center frequency, inputting the intermediate frequency signal into the bandpass filter to output a filtered signal through the bandpass filter.

[0074] Wherein, the bandwidth refers to the difference between the upper and lower limits, and the center frequency refers to the average of the upper and lower limits.

[0075] S2. Extract the distance-velocity matrix of the filtered signal from the motion detection channel, and generate the motion marker in the forward space through the motion detection channel and the distance-velocity matrix.

[0076] This invention extracts the distance-velocity matrix of the filtered signal from the motion detection channel, and generates a motion marker in the foreground space using the motion detection channel and the distance-velocity matrix. This allows for rapid response to moving targets, retains the real-time advantages of traditional solutions, and reduces the false alarm rate in dynamic scenes.

[0077] In one embodiment of the present invention, extracting the distance-velocity matrix of the filtered signal in the motion detection channel includes: inputting the filtered signal into the FFT processor of the motion detection channel; performing a one-dimensional FFT on the filtered signal within a single frequency modulation cycle in the FFT processor to obtain a distance-dimensional spectrum within the single frequency modulation cycle; performing a slow-time FFT on the filtered signal within the same distance cell in multiple consecutive frequency modulation cycles to obtain a velocity-dimensional spectrum within the same distance cell; and arranging the distance-dimensional spectrum and the velocity-dimensional spectrum into a distance-velocity matrix.

[0078] The FFT processor refers to hardware or application-specific integrated circuits (ASICs) specifically designed for fast calculation of the Fast Fourier Transform (FFT). Its core function is to efficiently convert time-domain signals into frequency-domain signals. The distance-dimensional spectrum refers to the spectrum obtained after performing an FFT on the filtered signal within a single frequency modulation cycle. The horizontal axis represents frequency, and the vertical axis represents amplitude. The frequency value is proportional to the target distance and is used to extract target distance information. The velocity-dimensional spectrum refers to the spectrum obtained by performing another FFT (slow-time FFT) on the filtered signal sequence of the same distance unit within multiple consecutive frequency modulation cycles. The horizontal axis represents Doppler frequency, and the vertical axis represents amplitude. The frequency value is proportional to the target radial velocity and is used to extract target velocity information. The distance-velocity matrix is ​​a two-dimensional matrix formed by arranging the distance-dimensional spectrum of several frequency modulation cycles into rows by distance units and arranging the velocity-dimensional spectrum of the corresponding distance units into columns. The matrix elements carry both distance and velocity information for subsequent detection.

[0079] Optionally, the extraction of the range-velocity matrix of the filtered signal in the motion detection channel is specifically as follows: In the FMCW radar, the relationship between the intermediate frequency signal frequency f_beat and the target distance R is: R = (c f_beat) / (2S), where c is the speed of light and S is the frequency modulation slope (Hz / s). Therefore, f_beat on the frequency axis of the range dimension spectrum is converted into distance using the above formula. This distance is arranged in rows according to the range units. The range dimension spectrum itself does not directly indicate the distance, but each frequency point uniquely corresponds to a distance value. The same range unit is sampled over multiple chirp (i.e., frequency modulation cycle) periods to obtain a complex sequence. Then, an FFT (slow-time FFT) is performed to obtain the velocity dimension spectrum. The relationship between the Doppler frequency f_d in the velocity dimension spectrum and the radial velocity v is: v = (λf_d) / 2, where λ is 24. The wavelength is GHz, so the Doppler frequency axis can be converted into velocity. Similarly, the spectrum itself does not directly indicate velocity, but each frequency point uniquely corresponds to a velocity value. Arrange the velocity-dimensional spectrum of the corresponding distance unit into columns to obtain the distance-velocity matrix. That is, the row index of the distance-velocity matrix corresponds to the distance unit number, and the column index of the distance-velocity matrix corresponds to the velocity unit number.

[0080] In one embodiment of the present invention, generating motion markers in the forward space through the motion detection channel and the distance-velocity matrix includes: determining, through a hardware comparator in the motion detection channel, whether there is a velocity value in the distance-velocity matrix greater than a preset velocity threshold; when there is a velocity value in the distance-velocity matrix greater than the preset velocity threshold, using a high level as a motion marker on the distance unit corresponding to the velocity value greater than the preset velocity threshold in the forward space; and using a low level as a motion marker on the distance unit corresponding to the velocity value in the distance-velocity matrix that is not greater than the preset velocity threshold in the forward space.

[0081] The hardware comparator refers to a hardware circuit composed of a high-speed analog comparator or digital comparator unit. Its input terminal receives a voltage or digital code representing the speed value, and the other input terminal is connected to a reference voltage / digital threshold corresponding to 0.2 m / s. When the input speed value is greater than the threshold, it outputs a high level; otherwise, it outputs a low level, directly forming a motion indicator. Regarding the source of 0.2 m / s, in a typical office scenario, based on 200 sets of actual measurements, 0.2 m / s (i.e., the preset speed threshold) is set as the lower limit that can reliably distinguish between human micro-movements / breathing and environmental noise. In other scenarios, it is necessary to continuously set different preset speed thresholds to determine whether human micro-movements / breathing can be reliably distinguished from environmental noise. If the distinction is higher than 95%, other preset speed thresholds can be set in other scenarios. This will not be elaborated here. It should be noted that the value corresponding to a high level is 1, and the value corresponding to a low level is 0.

[0082] S3. Input the distance-velocity matrix into the vital signs channel to output the respiratory harmonic energy in the space in front through the vital signs channel, wherein the vital signs channel includes a static target window selector, a phase difference decomposer and a harmonic energy extractor.

[0083] This invention addresses the limitation of static human body micro-motion signals being unable to be separated from environmental vibrations by outputting respiratory harmonic energy in the space in front of the body through the vital signs channel. The phase difference decomposer accurately extracts 0.1-0.5Hz vital harmonic energy, breaking through the bottleneck of static human body detection.

[0084] In one embodiment of the present invention, the step of outputting respiratory harmonic energy in the forward space through the vital signs channel includes: acquiring a static target window selector, a phase difference decomposer, and a harmonic energy extractor in the vital signs channel; using the static target window selector to filter target range cells with zero velocity values ​​in the range-velocity matrix; using the phase difference decomposer to decompose the target range cells into in-phase and quadrature components; calculating the instantaneous phase corresponding to the in-phase and quadrature components; performing differential processing on the instantaneous phase to obtain the phase difference; and performing bandpass filtering processing on the phase difference within the target frequency band in the harmonic energy extractor to obtain the respiratory harmonic energy.

[0085] As another implementation, calculating the instantaneous phase corresponding to the in-phase component and the quadrature component includes: calculating the instantaneous phase corresponding to the in-phase component and the quadrature component using the following formula:

[0086]

[0087] in, Indicates instantaneous phase, Indicates in-phase components, This represents orthogonal components.

[0088] The static target window selector refers to a hardware logic unit used to traverse the range-velocity matrix row by row (range cell). When the velocity value corresponding to a certain row is zero, the index value of that row is latched and output as a subsequent processing window. The phase difference decomposer refers to an analog front-end circuit consisting of two completely symmetrical mixing branches and low-pass filters for each branch. The first branch outputs an orthogonal component Q through a 90-degree phase shifter and a first multiplier, and the second branch outputs an in-phase component I through a 0-degree phase shifter and a second multiplier. It should be noted that the input data for both the first and second branches are complex sampling sequences. This complex sampling sequence refers to the set of complex sampling sequences used in the step of performing slow-time FFT processing on the filtered signal within the same range cell in multiple consecutive frequency modulation cycles. That is, after each chirp cycle is processed by one-dimensional FFT, the column of complex values ​​falling on the same target range cell. The outputs of the first and second branches are respectively filtered by low-pass filters to form the final output. The I and Q baseband signals are used for subsequent phase calculations. That is, in the 0-degree branch, the complex sampling sequence is mixed in phase with the local 24GHz local oscillator (mixing is the process of a multiplier), and after low-pass filtering, the in-phase component I is obtained. In the 90-degree branch, the complex sampling sequence first passes through a 90-degree phase shifter to generate a quadrature local oscillator, then is mixed with the complex sampling sequence, and after low-pass filtering, the quadrature component Q is obtained. The harmonic energy extractor refers to a digital or analog signal processing unit, with its input port connected to a phase differential signal. The unit integrates a 0.1-0.5Hz bandpass filter and an energy accumulator. The output port provides the energy value within this frequency band. Regarding the source of 0.1-0.5Hz, it is as follows: the resting respiratory rate of an adult is about 12-20 breaths / minute, corresponding to the 0.2-0.5Hz frequency band. Considering individual differences and extremely slow breathing, the range is extended downwards to 0.1Hz. Therefore, the passband of the bandpass filter is set to 0.1-0.5Hz to cover all possible respiratory signal frequency bands.

[0089] See Figure 2 The image shown is a noise spectrum comparison diagram of a smart home anti-interference human presence detection method based on 24GHz radar provided in an embodiment of the present invention. Figure 2 In the diagram, the upper line represents the air conditioner vibration spectrum, with 38% of the energy in the 0.1-0.5 Hz range. The lower line represents the human respiratory spectrum, with 82% of the energy in the 0.1-0.5 Hz range. The 38% and 82% are obtained by dividing the area of ​​the curve in the 0.1–0.5 Hz range by the total area of ​​the entire curve and then multiplying by 100%. The vertical axis represents the dimensionless normalized power, indicating the relative energy at each frequency point, used to calculate the energy percentage within the range.

[0090] S4. Sample the noise spectrum baseline in the space in front, and convert the noise spectrum baseline into a noise interference flag through a frequency domain mutual exclusion verifier.

[0091] This invention addresses the problem of misjudgment caused by the overlap of mechanical vibration harmonics and respiratory signal frequency bands by sampling the noise spectrum baseline in the space in front and converting the noise spectrum baseline into a noise interference flag through a frequency domain mutual exclusion verifier. The dynamic calibration of the noise baseline and the frequency band energy ratio threshold can achieve frequency domain decoupling between vibration interference and vital signs.

[0092] In one embodiment of the present invention, sampling the noise spectrum baseline in the foreground space includes: when there is no one in the foreground space, using a 24GHz FMCW radar front-end module to collect the unmanned intermediate frequency signal in the foreground space; and performing spectrum analysis on the unmanned intermediate frequency signal to generate the noise spectrum baseline in the foreground space.

[0093] The unmanned intermediate frequency signal refers to the intermediate frequency signal when no one is present, which is consistent with the acquisition principle of S1 mentioned above, and will not be elaborated here.

[0094] Optionally, the process of performing spectrum analysis on the unmanned intermediate frequency signal to generate the noise spectrum baseline in the foreground space specifically involves: sending the unmanned intermediate frequency signal into the aforementioned FFT processor, performing an FFT once to obtain the frequency domain amplitude spectrum, and storing or caching the amplitude spectrum as the noise spectrum baseline for use by the subsequent frequency domain mutual exclusion verifier.

[0095] In one embodiment of the present invention, converting the noise spectrum baseline into a noise interference flag using a frequency domain mutual exclusion verifier includes: calculating the energy percentage of the noise spectrum baseline in the target frequency band in the frequency domain mutual exclusion verifier; outputting a high level to generate a noise interference flag when the energy percentage is not less than a preset percentage; and outputting a low level to generate a noise interference flag when the energy percentage is less than the preset percentage.

[0096] The target frequency band is consistent with the target frequency band mentioned above, which is 0.1-0.5Hz. The preset proportion is set to 5% in the office scenario. Regarding the source of the 5%, in 200 sets of office tests, when the energy proportion of environmental noise in the 0.1-0.5Hz frequency band is less than 5%, the false alarm rate of static human body detection is no more than 5%. When it is not less than 5%, the false alarm rate increases significantly. Therefore, this empirical value is used as the threshold.

[0097] S5. Based on the motion marker, the respiratory harmonic energy, and the noise interference marker, detect whether there is a human body in the space in front, and obtain the detection result of the presence of a human body.

[0098] This invention addresses the shortcomings of fragmented motion and static detection logic and poor environmental adaptability by using the motion marker, respiratory harmonic energy, and noise interference marker. The motion marker is prioritized for triggering and mutually exclusive with static vital signs verification, thereby comprehensively improving detection reliability.

[0099] In one embodiment of the present invention, the step of detecting whether a human body exists in the space in front based on the motion marker, the respiratory harmonic energy, and the noise interference marker, and obtaining a human body presence detection result, includes: when the motion marker is high, taking the presence of a dynamic human body in the space in front as a human body presence detection result; when the motion marker is low, if the respiratory harmonic energy is greater than a hardware-defined value and the noise interference marker is low, taking the presence of a static human body in the space in front as a human body presence detection result; when the motion marker is low, if the respiratory harmonic energy is not greater than a hardware-defined value, taking the absence of a human body in the space in front as a human body presence detection result; when the motion marker is low, if the noise interference marker is not low, taking the absence of a human body in the space in front as a human body presence detection result.

[0100] Specifically, the source of the hardware-fixed value is as follows: the office scene is set to 0.7. The setting principle is similar to the principle mentioned above, which states that in 200 sets of office tests, when the energy proportion of environmental noise in the 0.1-0.5Hz frequency band is less than 5%, the false alarm rate of static human detection is no more than 5%. When it is not less than 5%, the false alarm rate increases significantly. Therefore, this empirical value is used as the threshold. That is, through 200 actual tests and calibrations, the hardware-fixed value can ensure that the detection rate is not less than 95%.

[0101] See Figure 3 The diagram shown is a system architecture diagram of a smart home anti-interference human presence detection method based on 24GHz radar provided in an embodiment of the present invention. Figure 3 In this system, the dual-mode decision circuit receives three inputs (motion markers, respiratory energy, and noise markers) and makes decisions on the received data to generate a human presence detection result.

[0102] Furthermore, to better understand the comparison of static human detection performance in the above-mentioned method for anti-interference human presence detection in smart homes based on 24GHz radar, please refer to Table 1 below, which is a comparison of static human detection performance in an embodiment of the present invention for implementing the method for anti-interference human presence detection in smart homes based on 24GHz radar.

[0103]

[0104] Table 1

[0105] As can be clearly seen from Table 1 above, the method for detecting human presence in smart homes based on 24GHz radar can improve the detection rate and reduce the false alarm rate, ultimately achieving a technical effect that combines the reliability and anti-interference of static human detection.

[0106] Furthermore, to better understand the comparison of static human detection performance in the above-mentioned method for anti-interference human presence detection in smart homes based on 24GHz radar, please refer to Table 1 below, which is a comparison of static human detection performance in an embodiment of the present invention for implementing the method for anti-interference human presence detection in smart homes based on 24GHz radar.

[0107]

[0108] Table 2

[0109] As can be clearly seen from Table 1 above, in the method of anti-interference human presence detection in smart homes based on 24GHz radar, by setting the hardware fixed value to 0.7, the detection rate can be guaranteed to be no less than 95% for most occasions.

[0110] Compared to the problems described in the background technology, this embodiment of the invention uses a 24GHz FMCW radar front-end module integrated within a smart home to collect intermediate frequency signals in the space in front of the smart home. The intermediate frequency signals are then filtered to obtain a filtered signal, thus solving the problem of high-frequency interference in the original signal causing distortion in feature extraction. Simultaneously, bandpass filtering suppresses out-of-band noise, improves the signal-to-noise ratio, and provides a clean signal source for dual-channel processing. This embodiment extracts the range-velocity matrix of the filtered signal in the motion detection channel, and uses the motion detection channel and the range-velocity matrix to generate motion markers in the space in front, enabling rapid response to moving targets while retaining the real-time advantages of traditional solutions and reducing the false negative rate in dynamic scenes. This embodiment outputs the call signs in the space in front through the vital signs channel. This invention addresses the limitation of static human body micro-motion signals being unable to be separated from environmental vibrations by absorbing harmonic energy. A phase differential decomposer precisely extracts 0.1-0.5Hz vital harmonic energy, overcoming the bottleneck of static human body detection. In this embodiment, the noise spectrum baseline in the foreground space is sampled, and a frequency domain mutual exclusion verifier converts this baseline into a noise interference flag. This resolves the misjudgment caused by the overlap of mechanical vibration harmonics and respiratory signal frequency bands. Dynamic calibration of the noise baseline and frequency band energy ratio thresholds achieve frequency domain decoupling of vibration interference and vital signs. Furthermore, this invention addresses the shortcomings of fragmented motion and static detection logic and poor environmental adaptability by using the motion flag, respiratory harmonic energy, and noise interference flag. The motion flag prioritizes triggering and mutually excludes static vital signs verification, comprehensively improving detection reliability. Therefore, this invention achieves both high reliability and strong anti-interference capabilities in static human body detection.

[0111] like Figure 4 The diagram shown is a functional block diagram of a smart home anti-interference human presence detection system based on 24GHz radar according to the present invention.

[0112] The intelligent home anti-interference human presence detection system 400 based on 24GHz radar described in this invention can be installed in electronic devices. Depending on the functions implemented, the intelligent home anti-interference human presence detection system based on 24GHz radar may include a signal filtering module 401, a flag generation module 402, an energy output module 403, a noise conversion module 404, and a human detection module 405. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0113] In this embodiment of the invention, the functions of each module / unit are as follows:

[0114] The signal filtering module 401 is used to collect intermediate frequency signals in the space in front of the smart home based on the 24GHz FMCW radar front-end module integrated inside the smart home, and to filter the intermediate frequency signals to obtain a filtered signal.

[0115] The marker generation module 402 is used to extract the distance-velocity matrix of the filtered signal in the motion detection channel, so as to generate the motion marker in the front space through the motion detection channel and the distance-velocity matrix.

[0116] The energy output module 403 is used to input the distance-velocity matrix into the vital signs channel, so as to output the respiratory harmonic energy in the front space through the vital signs channel. The vital signs channel includes a static target window selector, a phase difference decomposer and a harmonic energy extractor.

[0117] The noise conversion module 404 is used to sample the noise spectrum baseline in the front space and convert the noise spectrum baseline into a noise interference flag through a frequency domain mutual exclusion verifier.

[0118] The human body detection module 405 is used to detect whether there is a human body in the space in front of it based on the motion marker, the respiratory harmonic energy and the noise interference marker, and to obtain the detection result of the presence of a human body.

[0119] In detail, the modules in the smart home anti-interference human presence detection system 400 based on 24GHz radar described in this embodiment of the invention employ the same methods as described above. Figure 1 The method described above uses the same technology as the 24GHz radar-based smart home anti-interference human presence detection method, and can produce the same technical effect, so it will not be repeated here.

[0120] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0121] Finally, it should be noted that in the above embodiments, each embodiment can be combined with each other or independent. Deleting any one of them will not affect the technical implementation of other embodiments. The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart home anti-interference human presence detection method based on 24GHz radar, characterized in that, The method includes: Based on the 24GHz FMCW radar front-end module integrated inside the smart home, the intermediate frequency signal in the space in front of the smart home is collected, and the intermediate frequency signal is filtered to obtain a filtered signal. The distance-velocity matrix of the filtered signal is extracted in the motion detection channel, and the motion marker in the forward space is generated by the motion detection channel and the distance-velocity matrix. The distance-velocity matrix is ​​input into the vital signs channel to output the respiratory harmonic energy in the space in front through the vital signs channel. The vital signs channel includes a static target window selector, a phase difference decomposer, and a harmonic energy extractor. The noise spectrum baseline in the foreground space is sampled, and the noise spectrum baseline is converted into a noise interference flag using a frequency domain mutual exclusion verifier. Specifically, this includes: In the frequency domain mutual exclusion verifier, the energy percentage of the noise spectrum baseline within the target frequency band is calculated; When the energy percentage is not less than the preset percentage, a high level is output to generate a noise interference flag; When the energy percentage is less than a preset percentage, a low level is output to generate a noise interference flag; Based on the motion marker, the respiratory harmonic energy, and the noise interference marker, the presence of a human body in the space in front is detected, and the presence of a human body is obtained, specifically including: When the motion indicator is high, the presence of a dynamic human body in the space in front is taken as the detection result of the presence of a human body; When the motion indicator is low, if the breathing harmonic energy is greater than the hardware-fixed value and the noise interference indicator is low, the presence of a static human body in the space in front is taken as the detection result of the presence of a human body. When the motion indicator is low, if the breathing harmonic energy is not greater than the hardware-fixed value, the absence of a human body in the space in front is taken as the detection result of the presence of a human body. When the motion indicator is low, if the noise interference indicator is not low, the absence of a human body in the space in front is taken as the detection result of the presence of a human body.

2. The smart home anti-interference human presence detection method based on 24GHz radar as described in claim 1, characterized in that, The 24GHz FMCW radar front-end module integrated within the smart home collects intermediate frequency signals from the space in front of the smart home, including: The voltage-controlled oscillator, power amplifier, transmitting antenna, receiving antenna, and mixer in the 24GHz FMCW radar front-end module are obtained, wherein the mixer includes a first input terminal, a second input terminal, and an intermediate frequency output terminal; A 24 GHz frequency-modulated continuous wave with a linear frequency variation over time is generated by the voltage-controlled oscillator. The 24 GHz frequency-modulated continuous wave is amplified by the power amplifier to obtain an amplified continuous wave; The amplified continuous wave is radiated into the space in front via the transmitting antenna; The received antenna captures the reflected echo signal of the amplified continuous wave in the space in front; The reflected echo signal in the receiving antenna is received using the first input terminal; The second input terminal is used to receive the 24 GHz frequency-modulated continuous wave from the voltage-controlled oscillator; Based on the input-output coupling relationship between the first input terminal, the second input terminal and the intermediate frequency output terminal respectively, the reflected echo signal is mixed with the 24 GHz frequency-modulated continuous wave to form an intermediate frequency signal using the intermediate frequency output terminal.

3. The smart home anti-interference human presence detection method based on 24GHz radar as described in claim 1, characterized in that, The step of filtering the intermediate frequency signal to obtain a filtered signal includes: The bandwidth and center frequency of the bandpass filter are set based on the upper and lower limits of the intermediate frequency signal; After setting the bandwidth and the center frequency, the intermediate frequency signal is input into the bandpass filter to output a filtered signal through the bandpass filter.

4. The smart home anti-interference human presence detection method based on 24GHz radar as described in claim 1, characterized in that, The step of extracting the distance-velocity matrix of the filtered signal in the motion detection channel includes: The filtered signal is input into the FFT processor of the motion detection channel; In the FFT processor, the filtered signal is processed by one-dimensional FFT within a single frequency modulation cycle to obtain the distance-dimensional spectrum within the single frequency modulation cycle; The filtered signal is processed by slow-time FFT within the same distance cell over multiple consecutive frequency modulation cycles to obtain the velocity dimension spectrum within the same distance cell. Arrange the distance-dimensional spectrum and the velocity-dimensional spectrum into a distance-velocity matrix.

5. The smart home anti-interference human presence detection method based on 24GHz radar as described in claim 1, characterized in that, The step of generating motion markers in the forward space through the motion detection channel and the distance-velocity matrix includes: The hardware comparator in the motion detection channel determines whether there is a velocity value in the distance-velocity matrix that is greater than a preset velocity threshold. When there is a velocity value greater than a preset velocity threshold in the distance-velocity matrix, a high level is used as a motion marker on the distance cell corresponding to the velocity value greater than the preset velocity threshold in the forward space; The low level is used as the motion indicator on the distance cell in the space in front of the velocity value that is not greater than a preset velocity threshold in the distance-velocity matrix.

6. The smart home anti-interference human presence detection method based on 24GHz radar as described in claim 1, characterized in that, The output of respiratory harmonic energy in the space in front through the vital signs channel includes: Acquire the static target window selector, phase difference decomposer, and harmonic energy extractor in the vital signs channel; Use a static target window gating tool to filter target range cells with zero velocity values ​​in the range-velocity matrix; The target range cell is decomposed into in-phase and quadrature components using the phase difference decomposer. The instantaneous phases corresponding to the in-phase component and the quadrature component are calculated using the following formula: ； in, Indicates instantaneous phase, Indicates in-phase components, Indicates orthogonal components; The instantaneous phase is differentially processed to obtain the phase difference; In the harmonic energy extractor, the phase difference is subjected to bandpass filtering within the target frequency band to obtain breathing harmonic energy.

7. The smart home anti-interference human presence detection method based on 24GHz radar as described in claim 1, characterized in that, The sampling of the noise spectrum baseline in the foreground space includes: When there is no one in the space in front, the intermediate frequency signal of the unmanned person in the space in front is collected using the 24GHz FMCW radar front-end module. Spectral analysis is performed on the unmanned intermediate frequency signal to generate a noise spectrum baseline in the space in front.

8. A smart home anti-interference human presence detection system based on 24GHz radar, characterized in that, The system includes: The signal filtering module is used to collect intermediate frequency signals in the space in front of the smart home based on the 24GHz FMCW radar front-end module integrated inside the smart home, and to filter the intermediate frequency signals to obtain a filtered signal. The marker generation module is used to extract the distance-velocity matrix of the filtered signal in the motion detection channel, so as to generate the motion marker in the front space through the motion detection channel and the distance-velocity matrix; An energy output module is used to input the distance-velocity matrix into the vital signs channel to output the respiratory harmonic energy in the space in front through the vital signs channel. The vital signs channel includes a static target window selector, a phase difference decomposer, and a harmonic energy extractor. The noise conversion module is used to sample the noise spectrum baseline in the foreground space and convert the noise spectrum baseline into a noise interference flag through a frequency domain mutual exclusion verifier. Specifically, it includes: In the frequency domain mutual exclusion verifier, the energy percentage of the noise spectrum baseline within the target frequency band is calculated; When the energy percentage is not less than the preset percentage, a high level is output to generate a noise interference flag; When the energy percentage is less than a preset percentage, a low level is output to generate a noise interference flag; The human body detection module is used to detect whether a human body exists in the space in front of the device based on the motion marker, the respiratory harmonic energy, and the noise interference marker, and to obtain a human body presence detection result. Specifically, it includes: When the motion indicator is high, the presence of a dynamic human body in the space in front is taken as the detection result of the presence of a human body; When the motion indicator is low, if the breathing harmonic energy is greater than the hardware-fixed value and the noise interference indicator is low, the presence of a static human body in the space in front is taken as the detection result of the presence of a human body. When the motion indicator is low, if the breathing harmonic energy is not greater than the hardware-fixed value, the absence of a human body in the space in front is taken as the detection result of the presence of a human body. When the motion indicator is low, if the noise interference indicator is not low, the absence of a human body in the space in front is taken as the detection result of the presence of a human body.

Images