Method and device for eliminating random noise of radar acquisition signal in radar for biological signal measurement

Abstract

The present invention relates to a method and apparatus for effectively reducing and eliminating various vibration noises flowing in from the outside during the acquisition of a biological signal of a human body. The present invention includes: step (a), as a voltage-controlled oscillator (VCO) of an intermediate frequency (IF) signal generator generates an oscillation frequency with a specified period through a voltage (V(t)) applied from the outside and transmits it to a main part of a dynamic target through a single transmitting antenna, signals reflected by the dynamic target and various signals generated around the dynamic target are received through n receiving antennas; step (b), the IF signal generator generates respective Doppler intermediate frequency signals from the n received signals; step (c), the plurality of Doppler intermediate frequency signals input from the IF signal generator are respectively converted into digital data through an analog-to-digital converter; step (d), through a fast Fourier transformer, the digital signals acquired by the analog-to-digital converter (20) within a unit time are formed into a data string having symbols at a sampling time interval, and the data string is converted into a frequency component symbol string (symbol set) of a plurality of indexes (Index); step (e), the calculation unit of the signal calculator calculates a value obtained by adding the plurality of index symbols converted into the frequency component symbol string to each index set and dividing by the number of n receiving antennas; and step (f), for the value calculated by the calculation unit, the deviation of the spectral components of the co-occurring periodic signal and the non-periodic signal that does not co-occur is classified according to a specified reference threshold (Threshold) to filter only the periodic signal. The present invention effectively eliminates random noise contained in the biological signal region by using a Doppler radar for measuring biological signals composed of a single transmitting unit and a plurality of receiving units, thereby improving the accuracy of sensing biological signals in a non-contact manner.

Description

Technical Field

[0001] The present invention relates to a method and apparatus for effectively reducing and eliminating various vibration noises flowing in from the outside during the acquisition of biological signals of a human body by using a microwave Doppler radar. Background Art

[0002] Generally, as a kind of sensing technology, a radar can accurately measure the distance to an object and the relative speed of the object with respect to an observation point. A radar device operates by transmitting electromagnetic waves of microwaves to an object and receiving the electromagnetic waves reflected by the object. The processed signal is converted into a state that can be used by an operator or surrounding devices controlled by the radar. Technologies for detecting actions or biological signals using a radar are gradually becoming commercialized. Therefore, there are solutions for solving various noise problems in the radar received signal caused by possible external influences in the environment where a radar is actually used.

[0003] In addition, a Doppler Radar applies the Doppler effect of an electric wave and detects a moving target based on the difference between the transmission frequency and the reflection frequency of a radar wave transmitted toward the target. It is used in a weather radar, a self-navigation system device for an aircraft, and a military radar. In the case of being used for weather, it measures the change in wind speed generated inside a cloud. In the case of being used for a self-navigation system device, it calculates the current position by measuring the speed at which an electric wave reaches the ground. In the case of being used for a military radar, a pulse Doppler radar is mainly used, and this pulse Doppler radar generally captures and tracks only a target moving within a reflected wave on the ground and at sea level with a single pulse signal.

[0004] Generally, in Figure 1 the method of measuring biological signals using an existing microwave radar, the Doppler effect caused by the vibration or movement of a human body due to a heartbeat, breathing, or movement is sensed by collecting data.

[0005] However, during the process of measuring the biological signals of an actual human body, a Doppler effect caused by the movement of the human body, the movement of muscles, or surrounding environmental factors will additionally occur. Therefore, in Figure 2 in the acquisition signal of actual radar sensing, in addition to biological signal data, random noise is also loaded on the same regions in the time domain and frequency domain where biological signals exist. This random noise has the characteristic of changing in various ways over time due to the lack of periodicity, while on the contrary, biological signals have the characteristic of maintaining a specified periodic pattern. Therefore, research should be conducted to distinguish by calculating the difference from the random noise doped in the human heartbeat spectrum region.

[0006] As the prior art related to the present invention, the method for determining biological information disclosed in Patent Document 1 includes the following steps: generating a frame set by accumulating single frames in which radar pulses reflected from a target whose heartbeat is to be measured overlap according to a predetermined reception time; eliminating, for each single frame included in the generated frame set, the single frame in the transition interval where the maximum peak is generated due to the movement of the target, for the first sampling index representing the maximum peak and the second sampling index which is the sampling index corresponding to the maximum peak in the first sampling index; as the single frame in the transition interval where the maximum peak is generated due to the movement of the target is eliminated, for the frame set included in the blank interval, applying an algorithm for detecting a periodic pattern of incomplete data to identify the peak frequency to determine the heartbeat frequency of the target, the radar pulse reflected from the target is a radar signal reflecting the movement of the target, and the step of eliminating the part where the transition interval is generated includes the following steps: extracting, for each single frame included in the frame set, the first sampling index representing the maximum peak; determining the sampling index corresponding to the maximum peak with the largest number in the first sampling index representing the maximum peak as the second sampling index for generating the reference of the motion profile; generating a motion profile based on the movement of the target using the difference between the first sampling index and the second sampling index; and arranging the single frames included in the frame set using the motion profile.

[0007] In addition, Patent Document 2 discloses a multi-carrier Doppler radar. After transmitting a baseband signal, the Doppler is measured based on the baseband signal returned by being reflected by an object. The Doppler radar uses a baseband signal including multi-carriers as the baseband signal. The Doppler radar includes a signal processing unit that converts the received baseband signal into N frequency signals through a Fourier converter and selects any M frequency signals from the N converted frequencies, extracts phase values for the M frequency signals output by multiplying the selected M frequency signals by a calibration constant respectively, and applies a filter to calculate the output value for the M distance information output by multiplying the extracted M phase values by a distance conversion constant respectively.

[0008] Prior art documents

[0009] Patent documents

[0010] Patent Document 1: Korean Patent Gazette No. 10-1948386 (Publication Date: February 14, 2019)

[0011] Patent Document 2: Korean Patent Publication Gazette No. 10-2018-0010713 (Publication Date: January 31, 2018) Summary of the invention

[0012] Technical problem

[0013] In order to solve the above problems, an object of the present invention is to effectively eliminate random noise in a region of a biological signal with relative periodicity doped with random noise.

[0014] Moreover, another object of the present invention is to improve the accuracy of measuring a biological signal in a non-contact manner by a microwave radar.

[0015] Solution to the problem

[0016] To achieve the above object, a method for eliminating random noise of a radar acquisition signal in a biological signal measurement radar provided by the present invention includes: step (a), as a voltage-controlled oscillator (VCO) of an intermediate frequency signal generator generates an oscillation frequency with a specified period through an externally applied voltage V(t) and transmits it to a main part of a dynamic target through a single transmitting antenna, signals reflected by the dynamic target and various signals generated around the dynamic target are received through n receiving antennas; step (b), the intermediate frequency signal generator generates respective Doppler intermediate frequency signals from the n received signals; step (c), the plurality of Doppler intermediate frequency signals input from the intermediate frequency signal generator are respectively converted into digital data through an analog-to-digital converter; step (d), through a fast Fourier transformer, the digital signals collected by the analog-to-digital converter per unit time are formed into a data string having symbols at a sampling time interval and the data string is converted into a frequency component symbol string (symbol set) of a plurality of indexes (Index); step (e), the calculation unit of the signal calculator calculates a value obtained by adding the plurality of index symbols converted into the frequency component symbol string for each index set and dividing by the number of n receiving antennas; and step (f), for the value calculated by the calculation unit, the deviation of the spectral components of the co-occurring periodic signal and the non-periodic signal that does not co-occur is classified according to a specified reference threshold (Threshold) to filter only the periodic signal.

[0017] Moreover, the device for eliminating random noise in the radar acquisition signal in the biosignal measurement radar provided by the present invention includes: an intermediate frequency signal generator, which generates an oscillation frequency with a specified period as the voltage controlled oscillator (VCO) generates an oscillation frequency with a specified period through the externally applied voltage V(t) and transmits it through a single transmitting antenna Tx. The reflected waves reflected from the radar signal acquisition area including dynamic targets at a specified distance are respectively received by multiple receiving antennas Rx-1 to Rx-n, and a Doppler intermediate frequency signal is generated from the received n signals; an analog-to-digital converter for converting the n analog signals generated from the above intermediate frequency signal generator into digital signals; a signal calculator, which makes the digital signals acquired by the above analog-to-digital converter per unit time form a data string with symbols at a sampling time interval through a fast Fourier transformer and converts the formed data string into a frequency component symbol string (symbol set) of indexes (Index), and then calculates the value obtained by adding multiple index symbols to each index set and dividing by the number of n receiving antennas Rx; and a digital filtering signal processor, which classifies and filters the deviation of the spectral components of the co-occurring periodic signal and the non-periodic signal that does not co-occur according to a specified reference threshold (Threshold) for the value calculated by the above calculation unit.

[0018] Moreover, in the present invention, the above receiving antennas Rx-1 to Rx-n can be arranged at positions and in directions capable of receiving signals from dynamic targets and other areas simultaneously.

[0019] Moreover, in the present invention, the above digital filtering signal processor can filter according to a reference threshold through one of a band-pass filter (BPF), a high-pass filter (HPF), or a low-pass filter (LPF) with a specified frequency band.

[0020] Effects of the Invention

[0021] The present invention has the following effects, that is, in the case where random noise is doped in the biosignal area with relative periodicity, the non-contact biosignal measurement Doppler radar composed of a single transmitting part and multiple receiving parts can effectively reduce and eliminate random noise to improve the accuracy of sensing biosignals. Description of the Drawings

[0022] Figure 1 It is a block diagram showing the concept of a radar commonly used for measuring biosignals.

[0023] Figure 2 It is a graph showing the heartbeat spectrum and the heartbeat spectrum doped with random noise as biosignals respectively.

[0024] Figure 3Block diagram of an apparatus for eliminating random noise in a radar acquisition signal in a radar for biological signal measurement according to an embodiment of the present invention.

[0025] Figure 4 Structural diagram of an apparatus for eliminating random noise in a radar acquisition signal in a radar for biological signal measurement according to the present invention.

[0026] Figure 5 Flowchart of a method for eliminating random noise in a radar acquisition signal in a radar for biological signal measurement according to the present invention.

[0027] Figure 6 Table representing a frequency component symbol string of multiple indices by Fourier transform for an apparatus for eliminating random noise in a radar acquisition signal in a radar for biological signal measurement according to the present invention.

[0028] Figure 7 Graph of the Fourier transform result spectrum of each receiving section of a microwave Doppler radar in an apparatus for eliminating random noise in a radar acquisition signal in a radar for biological signal measurement according to the present invention. Detailed implementation mode

[0029] Hereinafter, an embodiment of an apparatus for eliminating random noise in a radar acquisition signal in a radar for biological signal measurement according to the present invention will be described with reference to the accompanying drawings.

[0030] In Figure 3 and Figure 4 The intermediate frequency (IF, intermediate frequency) signal generator 10 of the microwave Doppler radar 1 generates an intermediate frequency signal by the difference between the oscillation frequency generated by the voltage controlled oscillator 11 (VCO, Voltage Controlled Oscillator) and the received frequency. The intermediate frequency signal generator 10 includes a transmitting antenna for transmitting an oscillation frequency of a specified period output from the voltage controlled oscillator 11 to the outside. Moreover, the intermediate frequency signal generator 10 includes a receiving antenna for receiving the signal transmitted from the transmitting antenna, for example, receiving the signal reflected by a dynamic target at a specified distance, for example, a human body, and receiving signals from the radar signal acquisition area around the human body. Among them, the transmitting antenna Tx is single, and the receiving antennas Rx-1 to Rx-n are multiple. Preferably, the receiving antennas Rx-1 to Rx-n should be arranged at appropriate positions and in appropriate directions so as to be able to receive signals from the dynamic target and other areas simultaneously.

[0031] Moreover, in the intermediate frequency signal generator 10, mixers M1 to Mn are coupled to respective receiving antennas Rx-1 to Rx-n. The mixers M1 to Mn generate Doppler intermediate frequency signals based on the difference between the oscillation frequency output by the voltage controlled oscillator 11 and the n signals received by the respective receiving antennas Rx-1 to Rx-n.

[0032] The intermediate frequency signal generator 10 has the same structure as a Doppler radar transceiver for transmitting and receiving signals, including a single transmitting antenna Tx and a plurality of receiving antennas Rx-1 to Rx-n. Thus, in the intermediate frequency signal generator 10, as the voltage controlled oscillator 11 generates an oscillation frequency with a specified period based on an externally applied voltage V(t) and transmits it through the single transmitting antenna Tx, as reflected waves reflected from a radar signal acquisition area including a dynamic target at a specified distance are respectively received by the plurality of receiving antennas Rx-1 to Rx-n, Doppler intermediate frequency signals are generated in the n received signals.

[0033] The analog-to-digital converter 20 (Analog / Digital Converter) is used to convert the input analog signal into a digital signal for output, and converts the n analog signals generated by the intermediate frequency signal generator 10 into digital signals for output respectively. In this case, in order to convert the digital signal into a frequency symbol, fast Fourier transform needs to be performed. Therefore, data strings are formed at the sampling time intervals using the data of the digital signals collected per unit time.

[0034] The signal calculator 30 includes a fast Fourier transformer (FFT, Fast Fourier Transformer) and a calculation unit 32. The fast Fourier transformer 31 converts the data strings of the digital signals collected by the analog-to-digital converter 20 per unit time into symbols of frequency components and forms n spectral index data strings therefrom. Moreover, the calculation unit 32 is used to calculate the value obtained by adding the formed multiple spectral index symbols for each index set and dividing by the number of the n receiving antennas Rx-1 to Rx-n.

[0035] The digital filtering signal processor 40 classifies and filters the deviation of the spectral components of the co-occurring periodic signal and the non-co-occurring aperiodic signal according to a specified reference threshold value (Threshold) for the values calculated by the signal calculator 30 within a unit time. That is, although the calculation results of the co-occurring periodic signals are maintained at values of a specified magnitude, while the non-co-occurring aperiodic signals are reduced to very small values. Therefore, after significantly increasing the deviation of the spectral components of the periodic and aperiodic signals, the digital filtering signal processor 40 classifies and filters according to an appropriate reference threshold value to process the signals. As the filter for filtering according to the reference threshold value, a band-pass filter (BPF), a high-pass filter (HPF), a low-pass filter (LPF), etc. of a specified frequency band can be appropriately selected and used.

[0036] Hereinafter, a method for eliminating the random noise of the radar acquisition signal in the biological signal measurement radar of the present invention having the above structure will be described.

[0037] First, in Figure 5 , the voltage-controlled oscillator 11 of the intermediate-frequency signal generator 10 generates an oscillation frequency with a specified period from the externally applied voltage V(t) and transmits it to the main part of the dynamic target through the transmitting antenna Tx (step S1). Preferably, the signal transmitted from the transmitting antenna Tx should be transmitted toward the human body part that can most effectively collect the biological signal as the dynamic target.

[0038] The signals reflected by the dynamic target and various signals around the dynamic target are received by n receiving antennas Rx-1 to Rx-n (step S2). In this case, the n receiving antennas Rx-1 to Rx-n include: the receiving antenna Rx-1 for receiving the signal containing the most biological signals; and a plurality of receiving antennas Rx-2 to Rx-n for receiving biological signals and signals containing other actions of the human body or sound effects generated around.

[0039] The n signals received from the plurality of receiving antennas Rx-1 to Rx-n are respectively generated into Doppler intermediate-frequency signals by the intermediate-frequency signal generator 10 (step S3). Moreover, the analog-to-digital converter 20 converts the plurality of Doppler intermediate-frequency signals generated by the intermediate-frequency signal generator 10 and input thereto into respective digital data (step S4). During the period when the plurality of receiving antennas Rx-1 to Rx-n receive signals, in the time dimension, since the digital data continuously repeats, data can be continuously collected in real time according to the unit time.

[0040] As the digital signals acquired by the analog-to-digital converter 20 within a unit time are generated by the fast Fourier transformer 31 of the signal calculator 30 into a data string with symbols at the sampling time interval, the above data string is converted into frequency component symbols (symbol set) of multiple indexes (step S5).

[0041] On the other hand, the spectrum conversion of the Doppler intermediate frequency signal is achieved through the following process. That is, the spectrum conversion of the frequency components is to find the periodicity in the time-domain signals acquired within a specified time and convert it into frequency components. Therefore, digital sampling data in the time domain acquired within a specified time is required to implement the fast Fourier transform. And the symbol of the sampling time interval is the symbol of the original signal component sampled as a digital signal in the time domain. For example, if the time-domain signal data for 30 seconds is acquired and Fourier transform is implemented, the spectrum representing the frequency components of all signals showing periodicity within 30 seconds will be presented. In this case, the horizontal axis on the spectrum curve becomes the frequency axis, and this is the spectrum index.

[0042] Multiple indexes are as Figure 6 shown in the table. Multiple received signals are converted into a string of frequency component symbols of I(m) by the fast Fourier transformer 31. Moreover, the calculation unit 32 adds the multiple index symbols obtained from the multiple receiving antennas Rx-1 to Rx-n respectively for each index I(1), I(2), I(3),...I(m) set and calculates the value divided by the number n of the receiving antennas Rx-1 to Rx-n (step S6).

[0043] Among them, the signals of the n receiving antennas Rx-1 to Rx-n are represented by n spectrum index data, and the calculation is performed by the index set of the same spectrum components as each antenna. Moreover, I(1), I(2),...I(m) represent the actual frequency (Hz) components, and their symbol data represents the spectrum intensity.

[0044] And, in Figure 7 the spectrum curve graph, although the co-occurring periodic signals maintain a specified value, the non-periodic signals that do not co-occur are reduced to a very small value. Therefore, a spectrum deviation can be generated between the periodic signals and the non-periodic signals. That is, the frequency-domain index I(m) set is calculated as I(1) = a1 + b1 + c1...(n)1, I(2) = a2 + b2 + c2...(n)2, I(3) = a3 + b3 + c3...(n)3, I(m) = a(m) + b(m) + c(m)…(n)(m), etc. If the number of the receiving antennas Rx-1 to Rx-n is n, the calculation results of each index are I(1) / n, i(2) / n, I(3) / n...I(m) / n.

[0045] Furthermore, the digital filter signal processor 40 classifies the deviation of the spectral components of the co-occurring periodic signal and the non-periodic signal that does not co-occur according to a threshold value of a specified reference for the value calculated by the calculation unit 32, and filters only the periodic signal (step S7). In this case, if classification is performed based on the threshold value, it is possible to leave only the values higher than the reference point, which increases the possibility of a biological signal that usually has a periodic component in the time domain.

[0046] Thus, the common received part can be extracted from the data received from each of the receiving antennas Rx-1 to Rx-n by the above method, and the non-common part can be filtered out as random noise and eliminated.

[0047] As described above, compared with the conventional method of measuring a biological signal using a Doppler radar for biological signal measurement composed of a single transceiver unit, the present invention has the following advantages, that is, it is possible to effectively eliminate the random noise included in the biological signal region by using a Doppler radar for biological signal measurement composed of a single transmitting unit and a plurality of receiving units, and improve the accuracy of sensing a biological signal in a non-contact manner.

[0048] In the above description, although the present invention has been described with reference to specific embodiments and the drawings, it is obvious to those of ordinary skill in the art to which the present invention pertains that various modifications and changes can be made without departing from the inventive concept and scope of the claimed invention.

Claims

1. A method for eliminating random noise in a radar acquisition signal in a radar for biological signal measurement, characterized in that, Comprising: Step (a), as the voltage-controlled oscillator of the intermediate-frequency signal generator generates an oscillation frequency with a specified period through an externally applied voltage (V(t)) and transmits it to the main part of the dynamic target through a single transmitting antenna, signals reflected by the above-mentioned dynamic target and various signals generated around the dynamic target are received by n receiving antennas; Step (b), the above-mentioned intermediate-frequency signal generator generates respective Doppler intermediate-frequency signals from the n received signals; Step (c), the multiple Doppler intermediate-frequency signals input from the above-mentioned intermediate-frequency signal generator are respectively converted into digital data by an analog-to-digital converter; Step (d), through a fast Fourier transformer, the digital signals collected by the above-mentioned analog-to-digital converter (20) within a unit time are formed into a data string with symbols at a sampling time interval, and the above-mentioned data string is converted into a symbol string of frequency components of multiple indices; Step (e), the calculation unit of the signal calculator calculates the value obtained by adding the multiple index symbols converted into the symbol string of frequency components to each index set and dividing by the number of n receiving antennas; And Step (f), for the value calculated by the above-mentioned calculation unit, the deviation of the spectral components between the co-occurring periodic signal and the non-periodic signal that does not co-occur is classified according to a specified reference threshold to filter only the periodic signal.

2. A device for eliminating random noise in a radar acquisition signal in a biological signal measurement radar, characterized in that, Comprising: An intermediate-frequency signal generator, as the voltage-controlled oscillator generates an oscillation frequency with a specified period through an externally applied voltage (V(t)) and transmits it, the reflected waves reflected from the radar signal acquisition area including the dynamic target at a specified distance are respectively received by multiple receiving antennas (Rx-1 to Rx-n), and Doppler intermediate-frequency signals are generated from the n received signals; An analog-to-digital converter for converting the n analog signals generated from the above-mentioned intermediate-frequency signal generator into digital signals; A signal calculator, through a fast Fourier transformer, forms the digital signals collected by the above-mentioned analog-to-digital converter within a unit time into a data string with symbols at a sampling time interval, and after converting the formed data string into a symbol string of frequency components of indices, calculates the value obtained by adding the multiple index symbols to each index set and dividing by the number of n receiving antennas (Rx); And A digital filtered signal processor, for the value calculated by the above-mentioned signal calculator, classifies and filters the deviation of the spectral components between the co-occurring periodic signal and the non-periodic signal that does not co-occur according to a specified reference threshold.

3. The device for eliminating random noise of a radar acquisition signal in a radar for biological signal measurement according to claim 2, wherein The above-mentioned receiving antennas (Rx-1 to Rx-n) are arranged at positions and in directions capable of receiving signals from the dynamic target and other areas simultaneously.

4. The device for eliminating random noise of a radar acquisition signal in a radar for biological signal measurement according to claim 2, characterized in that, The above-mentioned digital filtered signal processor filters according to the reference threshold through one of a band-pass filter, a high-pass filter, or a low-pass filter with a specified frequency band.

Images