Signal analysis method and radar system

Abstract

To provide a radar system having high distance resolution, and high distance measurement accuracy in close ranges, regardless of chirp bandwidth.SOLUTION: A radar system is for measuring a distance from transmitting means 1 to a measurement object or movement speed. The radar system includes: the transmitting means 1 that generates a chirp signal whose frequency periodically increases and decreases with a constant modulation width and radiates the chirp signal into space as a traveling wave; receiving means 2 that detects the traveling wave reflected by the measurement object as a reflected wave and generates a beat signal from the reflected wave and the chirp signal; signal shaping means 4 that generates a shaped signal by removing distortion components from the beat signal generated by the receiving means 2; and signal analysis means 5 that derives the frequency of the reflected wave reflected by the measurement object from the shaped signal.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a signal analysis method and a radar system capable of measuring the distance to an object to be measured (hereinafter referred to as a "target") in a non-contact manner with high resolution and high accuracy. [Background technology]

[0002] 2. Description of the Related Art Conventionally, radar systems such as FMCW (Frequency Modulated Continuous Wave) radars are known as distance measurement systems that utilize microwaves, millimeter waves, and the like. FMCW radar transmits a frequency-swept continuous wave (hereafter referred to as a "chirp signal"), detects a beat signal generated by the transmitted signal and the reflected signal, and analyzes the beat signal to measure the distance to a target or its speed of movement. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2007-093576 A [Patent Document 2] International Publication No. 2009 / 138056 Summary of the Invention [Problem to be solved by the invention]

[0004] The range resolution of FMCW radar generally depends on the frequency of the chirp signal, which is determined by the inverse of the chirp bandwidth, and is said to have a range resolution equivalent to the theoretical value derived from the speed of light and the chirp bandwidth. However, in reality, the accuracy of distance measurement deteriorates due to distortion of the beat signal caused by coupling between the transmitting and receiving antennas, and the commonly used analysis method cannot identify the frequency or distance with a resolution equivalent to the theoretical value. For example, the Fast Fourier Transform (FFT) is generally used to analyze beat signals, but with FFT, problems arise such as the presence of side lobes and omissions (jumps) in the modulation frequency domain, resulting in a long minimum detection distance, making it difficult to distinguish and classify targets in close range, and to measure their distance and movement speed.

[0005] In addition, because the minimum unit of frequency bin (range bin) in the Discrete Fourier Transform (DFT) depends on the frequency of the chirp signal, it is not possible to accurately measure distances outside the distances corresponding to integer multiples of the frequency. As a result, the measurements of the distance to the target and the target's movement speed derived from this are far below the theoretically derived levels. For example, when detecting biological information, the distance to the target is generally within 1m to several meters, and the displacement of the target's body surface is generally 1 to 50mm for breathing and 0.1mm to 0.5mm for heartbeat. Moreover, when applying to detection of biological information in real life, it is often in a high-noise environment, and various noises are superimposed on the measurement data, making it extremely difficult to capture the biological information itself in a non-contact manner and measure it accurately.

[0006] Although the range resolution of FMCW radar can be improved by widening the chirp bandwidth, there is a problem that the efficiency of using the radio frequency band decreases. On the other hand, narrowing the chirp bandwidth not only worsens the range resolution, but also makes it difficult to measure distances at short distances.

[0007] The present disclosure has been made in consideration of the problems in the conventional technology described above, and aims to provide a radar system that has high distance resolution and high distance measurement accuracy at short distances, regardless of the chirp bandwidth. [Means for solving the problem]

[0008] A signal analysis method according to an embodiment of the present disclosure made to solve the above problems is a signal analysis method for measuring a distance and / or a motion speed to a measurement target in a radar system, the signal analysis method comprising the steps of: generating a chirp signal whose frequency periodically increases and decreases within a constant bandwidth, and radiating the chirp signal into space as a traveling wave; detecting the traveling wave reflected by the measurement target as a reflected wave, and generating a beat signal from the reflected wave and the chirp signal; generating a shaped signal by removing distortion components from the beat signal generated by the receiving means; and performing signal analysis of the shaped signal to derive a frequency component having a maximum amplitude as a specific frequency reflected by the measurement target. The signal analysis involves determining initial values ​​of each of the parameters such that a sum of squares of a difference between the shaped signal and a sinusoidal model signal represented by parameters of frequency f′, phase φ′, and amplitude A′ is minimized, and converging to an optimal solution of a nonlinear equation from each of the determined initial values, and the shaped signal is converted into a one-dimensional measurement value s(x n ) and the frequency f where the right hand side of equation (1) with the number of data points N is the minimum value. x The feature of this method is to obtain the amplitude A', phase φ' and amplitude A'.

number

[0009] The signal analysis involves determining initial values ​​for each of the parameters so that the sum of squares of the difference between the shaped signal and a sinusoidal model signal expressed as parameters of frequency f', phase φ', and amplitude A' is minimized, and converging to an optimal solution of a nonlinear equation from each determined initial value, wherein the steepest descent method is applied to frequency f' and phase φ' of the parameters to cause convergence, and then the Newton method incorporating a deceleration method is applied to cause convergence, and then the steepest descent method is applied to amplitude A' of the parameters to cause convergence, and then the Newton method incorporating a deceleration method is applied to cause convergence, and the shaped signal is converted into a one-dimensional measurement value s(x n ) and the frequency f where the right hand side of equation (1) with the number of data points N is the minimum value. x This signal analysis method is characterized by determining the amplitude A', phase φ' and amplitude A'.

[0010] A radar system according to one embodiment of the present disclosure made to solve the above problems is a radar system for measuring a distance to a measurement object and / or a movement speed, and is characterized by comprising: a transmitting means for generating a chirp signal whose frequency increases and decreases periodically within a constant bandwidth and radiating the chirp signal into space as a traveling wave; a receiving means for detecting the traveling wave reflected by the measurement object as a reflected wave and generating a beat signal from the reflected wave and the chirp signal; a signal shaping means for generating a shaped signal by removing distortion components from the beat signal generated by the receiving means; and a signal analysis means for performing signal analysis of the shaped signal and deriving a frequency component with a maximum amplitude as the specific frequency reflected by the measurement object. The transmitting means may be configured to generate a chirp signal having a triangular or sawtooth waveform that periodically increases and decreases in a certain frequency band, and to radiate the chirp signal into space as a traveling wave.

[0011] The signal analysis means determines initial values ​​of the parameters so that the sum of squares of the difference between the shaped signal and a sinusoidal model signal expressed as parameters of frequency f', phase φ', and amplitude A' becomes a minimum value, and converges to an optimal solution of a nonlinear equation from each determined initial value, and the shaped signal is a one-dimensional measured value s(x n ) and the frequency f where the right hand side of equation (1) with the number of data points N is the minimum value. x It is possible to employ a configuration for determining the amplitude A', phase φ' and amplitude A'.

[0012] The signal analysis means is a signal analysis means for determining initial values ​​of each of the parameters so that a sum of squares of a difference between the shaped signal and a sinusoidal model signal expressed as parameters of frequency f', phase φ', and amplitude A' becomes a minimum value, and for converging to an optimal solution of a nonlinear equation from each determined initial value, wherein the signal analysis means applies the steepest descent method to frequency f' and phase φ' among the parameters to cause convergence, and then applies the steepest descent method to amplitude A' among the parameters to cause convergence, and then applies the steepest descent method to amplitude A' among the parameters to cause convergence, and then applies the Newton method incorporating the deceleration method to cause convergence, and the shaped signal is converted into a one-dimensional measured value s(x n ) and the frequency f where the right hand side of equation (1) with the number of data points N is the minimum value. x It is also possible to employ a configuration in which the amplitude A', phase φ' and amplitude A' are obtained.

[0013] The signal shaping means may be configured to generate the shaped signal by linearization processing using the non-negative least squares method for a chirp time zone in the beat signal, or may be configured to include a distortion removal means that performs a distortion removal process to remove a distortion region in a chirp time zone in the beat signal as an invalid analysis section, a distortion correction means that corrects remaining distortion in the beat signal after the invalid analysis section has been removed, and / or a waveform application means that generates a shaped signal by connecting a waveform that approximates an ideal waveform with suppressed distortion to a missing portion caused by the distortion removal process. In addition, a configuration can be adopted in which a calculation means is provided for calculating the distance from the transmitting means to the object to be measured from the specific frequency derived by the signal analysis means, and the frequency derived by the signal analysis means or the distance derived by the calculation means is configured to detect the presence of biological fluctuations as biological information. Effect of the Invention

[0014] According to a radar system according to an embodiment of the present disclosure, distortion of the beat signal caused by coupling between the transmitting and receiving antennas is eliminated, and by adopting an NHA, the distance resolution can be improved even with a narrow chirp bandwidth, thereby improving the ranging accuracy at close ranges. This makes it possible to accurately detect the presence or absence of nearby targets and the minute, weak movements of each target, and also enables selective monitoring of only targets (people) in a specific direction, which will contribute to the development of autonomous driving systems, driving support systems, and medical and nursing support systems. [Brief description of the drawings]

[0015] [Figure 1] FIG. 1 is a block diagram showing an outline of a radar system. [Diagram 2] 4 is a time chart showing an example of a beat signal used in a radar system. [Diagram 3] 4 is a flowchart showing an example of processing performed in the radar system. [Figure 4] 4A and 4B are a flowchart and a time chart showing an example of processing of a beat signal used in a radar system. [Diagram 5] A graph showing distance-frequency characteristics comparing signal analysis performed in a radar system with signal analysis using FFT. [Figure 6] FIG. 2 is a block diagram showing an example of means included in the radar system. [Figure 7] 4 is a time chart showing an example of a shaped signal that is subjected to signal analysis in a radar system. [Figure 8] 4 is a time chart showing an example of a signal analysis result using a conventional method (FFT). [Figure 9] 1 is a time chart showing an example of a signal analysis result obtained by a radar system (NHA). [Figure 10] FIG. 1 is a block diagram showing a hardware configuration of a portion of a radar system. [Figure 11] A time chart showing examples of (a) an FMCW frequency chirp, (b) an ideal waveform of a beat signal, (c) a distorted waveform superimposed on the beat signal, and (d) a detected waveform of the beat signal. [Figure 12] 4 is a time chart illustrating a beat waveform and a process of signal shaping processing. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, a radar system according to an embodiment of the present disclosure will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an outline of a radar system. This example comprises an oscillator unit 10 that outputs a nuclear power signal as a traveling wave, a transmitter unit 11 that transmits the traveling wave as radio waves toward a target 14, a receiver unit 12 that detects the wave reflected by the target 14 and outputs a mixed wave of the reflected wave and the nuclear power signal, and an analyzer unit 13 that analyzes the frequency components of the mixed wave.

[0017] <Radar system hardware configuration> The oscillator 10 outputs a continuous signal (chirp signal) whose frequency is swept within a specific bandwidth, for example, the frequency increases and decreases periodically with a constant modulation width. The oscillator unit 10 is composed of, for example, an oscillator (VCO: Voltage Controlled Oscillator) that generates a continuous signal maintaining a predetermined frequency, and a modulator (not shown) that applies frequency modulation to the signal output from the oscillator, and can output a signal with a series of different frequency components within a specific bandwidth.

[0018] The transmitting unit 11 is equipped with an amplifier such as a power amplifier (PA) 32 that amplifies the signal output from the oscillator 10, and a transmitting antenna (including an electrode) 34 for outputting the output of the amplifier as radio waves, and transmits the radio waves output from the oscillator 10 via the transmitting antenna 34 toward the target 14. The receiving unit 12 is composed of, for example, a receiving antenna 35 for receiving radio waves reflected by the target 14, a low noise amplifier (LNA) 33 for amplifying the radio waves (hereinafter referred to as the "received waves") received by the receiving antenna (including an electrode) 35, and a mixer 30 for generating a mixed wave of the chirp signal generated by the oscillator 10 and the received wave, or a mixed wave of the chirp signal and the output of the low noise amplifier 33 (hereinafter referred to as the "beat signal") (see FIG. 1). In the receiving unit 12, the chirp signal is split into two routes input to the mixer 30, and the phase of one of the chirp signals is shifted by 90 degrees using a phase-shift circuit 36 ​​to generate a mixed wave mixed with the in-phase component and a mixed wave mixed with the quadrature component.

[0019] [Analysis section configuration] The analysis section 13 controls the oscillation section 10, the transmission section 11 and the reception section 12, and also analyzes the frequency components of the beat signal generated by the reception section. The analysis unit 13 digitizes one or both of the signals I and Q received from the receiving unit 12, or digitizes both of them into complex signals, stores the digitalized data (hereinafter referred to as "time series data") in a readable manner, and analyzes the frequency components of the time series data. The analysis unit 13 is a part of the radar system and is composed of a computer system and the like. As shown in FIG. 10, the analysis unit 13 is equipped with a CPU (Central Processing Unit) 20 that controls each part, a read-only ROM (Read Only Memory) 21 that stores various information including various programs, a RAM (Random Access Memory) 22 that functions as a work area when various programs are executed and temporarily stores various information and reads out the stored various information, a storage device 23 that stores various information in a readable and writable manner, an input operation control device 24 that processes and controls input operations via a user interface, a signal input device 25 that inputs the received waves to be analyzed into the analysis unit 13 as analyzable data, and a display device 26 that displays various information.

[0020] The memory device 23 stores a control program that controls the transmission and reception of the radar system, and a signal analysis program (hereinafter referred to as the "system program") that analyzes and calculates signals detected by the radar system, as well as various other information including analysis target data such as frequency components of the signal that is the subject of signal analysis. For example, a hard disk or a non-volatile memory can be used as the storage device 23. The storage device 23 also includes a drive device that reads and writes various information from a storage medium such as a flexible disk or a memory card that is detachable from the main body. The various information stored in the storage device 23 is read out under the control of the CPU 20.

[0021] The input operation control device 24 receives an input operation via a predetermined operation device (not shown) serving as a user interface, such as a keyboard or a mouse, and supplies the CPU 20 with a control signal indicating the content of the operation. The display device 26 is, for example, a display device such as a liquid crystal display (LCD), and displays various information under the control of the CPU 20. For example, when a signal analysis program is started by the CPU 20, the display device 26 displays the screen thereof, and displays an input support screen, an operation screen, or the results of signal analysis, etc. The signal input device 25 includes an A / D converter that converts the beat signal output by the receiving unit 12 into time-series data and outputs the data to the analyzing unit 13 .

[0022] <System Functionality> By incorporating programs such as the system program into the hardware, this radar system functions as a transmitting means 1, a receiving means 2, a data recording means 3, a signal shaping means 4, a signal analysis means 5 and a calculation means 6 (see FIG. 6). With these functions, the generated chirp signal is emitted as a traveling wave to the target 14, and is distributed to a mixer 30 that generates a beat signal together with the reflected (received) signal from the target 14. Next, a discontinuous section of the chirp portion of the beat signal that has been filtered through a distortion removal filter 31 is extracted and subjected to NHA analysis, and the distance to the target 14 is detected from the analysis results (see FIG. 3 or FIG. 4). In this case, the "distance to the target" refers to the distance from the detection point of the reflected wave in the radar system to the target, and based on this "distance to the target", the distance from a desired position in the radar system can also be calculated.

[0023] The transmitting means 1 generates a chirp signal in the oscillator 10 whose frequency increases and decreases periodically with a constant modulation width, and radiates the chirp signal into space as a traveling wave using the transmitting unit 11.The transmitting unit 1 generates a triangular or sawtooth chirp signal swept in a band corresponding to the event to be detected (e.g., 24 GHz band) and with a chirp bandwidth W (e.g., 200 MHz) corresponding to the event to be detected, and outputs the chirp signal into the air as a traveling wave after passing through the power amplifier 32 and transmitting antenna 34. The receiving means 2 causes the receiving section (detection point) 12 to detect the traveling wave reflected by the target 14 as a reflected wave, and performs phase detection using the chirp signal generated by the oscillator 10 and the reflected wave reflected by the stationary target 14 or the moving target 14, and outputs a beat signal corresponding to the distance to the target 14 or the distance between the targets 14. The beat signal can be made to have a frequency corresponding to the distance to the target 14 or the distance between the targets 14 by adjusting the chirp bandwidth and chirp time (see FIG. 2). The data recording means 3 stores the time series data of the beat signal in the storage device 23 .

[0024] The transmitting means 1 transmits, for example, a signal having a frequency f 0 ~f 0 +W band (chirp bandwidth W) and T U =T D =T triangular wave (see Fig. 11(a)), or T U =T and T D It outputs a traveling wave with a sawtooth waveform of =0 and a frequency that increases and decreases periodically. If the chirp time is T (s), the distance to the target 14 is x (m), and the speed of light is c (m / s), then the frequency of the beat signal is f fm is expressed by the following formula (10). The distance resolution Δx is ffm = 1 / T, the distance resolution Δx is given by the following formula (11).

[0025] FIG. 11(b) shows an ideal beat signal waveform when the distance to the target is x=Δx. When x=Δx, f fm = 1 / T, so (when x = nΔx, f fm =n / T), one period of the beat signal is output during one chirp time. FIG. 11(c) shows an example of a distorted waveform superimposed on a beat signal. Due to objects in the immediate vicinity of the FMCW radar antenna or coupling between the transmitting and receiving antennas 34, 35, radio waves leak from the transmitting antenna 34 to the receiving antenna 35, and a beat signal is output even when no target 14 is present. This is thought to be due to a beat signal being reflected from a virtual object present at a distance of x<Δx. In this way, when x<Δx, f fm <1 / T, so that a part of one period of the beat signal is output during one chirp time. FIG. 11(d) shows an actual beat signal waveform when the distance to the target is x=Δx. When the distance to the target is x=Δx, the actual beat signal has a waveform in which the above-mentioned distorted waveform is superimposed on the ideal beat signal.

[0026] The signal shaping means 4 generates a shaped signal by removing distortion components from the time-series data to be analyzed, and includes distortion removal means 4a that performs distortion removal processing to remove a distorted region in the chirp band from the time-series data of a beat signal on which a distorted waveform as shown in FIG. 11(d) is superimposed, waveform application means 4b that performs waveform application processing to generate a shaped signal by connecting missing parts resulting from the distortion removal processing so as to approximate an ideal waveform with suppressed distortion, and distortion correction means 4c that corrects the remaining distortion of the beat signal after the removal of the invalid analysis section.

[0027] The distortion removing means 4a removes radar-characteristic distortion in the time-series data of the beat signal by the following process. That is, first, the desired time series data is retrieved from the storage device 23, and data contained in an appropriate analysis window (hereinafter referred to as "analysis target data") is extracted. Next, a distortion removal process is performed in which invalid analysis intervals (such as jump intervals or missing intervals) containing harmful distortion components in the chirp portion of the triangular or sawtooth waveform of the analysis target data are cut out, and discontinuous intervals are cut out to extract valid analysis intervals. Examples of the portions to be removed as invalid analysis intervals include the vicinity of the start and end points of unit chirps in the beat signal obtained for each chirp time. At this time, only the valid analysis intervals may be used for signal analysis (see Figures 12(b) and (c)).

[0028] The waveform application means 4b performs a waveform application process in which a part or the whole of the valid analysis interval is applied to the missing portion of the analysis target data generated by the distortion removal process so as to approximate an ideal waveform that is connected to the valid analysis interval and has suppressed distortion. By performing such a waveform application process, it is possible to supplement the missing analysis window as an invalid analysis interval. The distortion correcting means 4c corrects the distortion of the waveform in the effective analysis interval that remains in the beat signal after the distortion removal by the distortion removing means 4a. The distortion correcting means 4c can employ removal processing such as removal of low frequency components by a digital high-pass filter or removal of trends by polynomial approximation.

[0029] As an embodiment of the signal shaping means 4, for example, a process can be performed in which a triangular wave up-flow section, which has less distortion in the data to be analyzed, is taken as a valid analysis section, and a triangular wave down-flow section is removed as an invalid analysis section, and the valid analysis section is filled in the missing section by swapping its time axis so as to form the target shape (see FIG. 7).

[0030] The signal shaping means 4 can also perform processing on a part of the data to be analyzed (such as the valid analysis section) or all of it to shape the data into a waveform with reduced distortion by, for example, removing low-frequency components using a digital high-pass filter or removing trends using a quadratic polynomial approximation (see FIG. 12(C)). In this way, by removing the distortion components from the analysis target data, the frequency resolution in the effective analysis interval is improved. During the signal shaping process, the signal shaping means 4 outputs the waveform of the data to be analyzed to the display device 26.

[0031] The signal analysis means 5 derives the frequency of the reflected wave reflected by the target 14 from the shaped signal, and performs signal analysis of the data to be analyzed (hereinafter, "shaped signal data") that has been subjected to signal shaping processes such as distortion removal processing and waveform application processing by the signal shaping means 4. The signal analysis means 5 includes data storage means 5a which performs a data storage process for storing the shaped signal data to be analyzed in the storage device 23, parameter calculation means 5b which reads out the shaped signal data from the storage device 23 and performs signal analysis of the shaped signal data, and analysis result output means 5c which performs an analysis result output process to output the analysis result based on the frequency f', the amplitude A', and the initial phase φ' calculated by the parameter calculation means 5b.

[0032] The parameter calculation means 5b of this system uses Non-Harmonic Analysis (hereinafter, referred to as "NHA"), see Japanese Patent No. 5590547, as a signal analysis method. Instead of finding the frequency parameters of the Fourier transform equation for a nonperiodic signal shown in Equation (2), the NHA finds the frequency f', amplitude A', and initial phase φ' that minimize the right-hand side of the nonlinear equation shown in Equation (3) as the optimal solution of the nonlinear equation expressed as the sum of squares of the difference between the analysis target signal x(n) related to the shaped signal data and a sine wave model signal represented by the phase using frequency f' and initial phase φ' and amplitude A' (parameter calculation process: nonnegative least squares method (LSM)). Note that in Equation (3), L is the frame length (analysis window length), and f s is the sampling frequency [Hz]. By employing the above-mentioned signal analysis method, the parameter calculation means 5b can eliminate the influence of distortion and accurately calculate the frequency f', amplitude A', and initial phase φ' regardless of the analysis window length (the analysis window can be flexibly set, for example, it can be less than one period, or a length other than an integer multiple of the period, or it can be at unequal intervals).

[0033]

number

[0034]

number

[0035] The parameter calculation means 5b calculates appropriate initial values ​​for the amplitude A', frequency f', and initial phase φ', and converges to an optimal solution from these initial values ​​using a nonlinear equation solution. The appropriate initial values ​​can be calculated by applying any existing method, such as any frequency transformation such as DFT or wavelet transformation, or filtering.

[0036] The parameter calculation means 5b first applies the steepest descent method to the frequency parameters f', φ' that constitute the phase of the sine wave model signal in the formula (3) to obtain the frequency parameter f m ',φ m ' is calculated using formulas (4) and (5). In addition, μ m is a weighting coefficient based on the so-called deceleration method, and takes a value between 0 and 1 as appropriate in order to make the cost function calculated by each recurrence formula a monotonically decreasing sequence. Frequency parameter f m ',φ m If we can obtain the frequency parameter A', which is a coefficient of the sinusoidal model signal in equation (3), we can uniquely obtain the frequency parameter A' by equation (6). m ' to converge.

[0037]

number

[0038]

number

[0039]

number

[0040] The parameter calculation means 5b repeats this series of calculations to cause the amplitude A', the frequency f', and the initial phase φ' to converge with high accuracy. In the NHA of the parameter calculation means 5b, the frequency parameters f', φ' constituting the phase of the sine wave model signal in the formula (3) and the frequency parameter A' as a coefficient are separately calculated, so that the calculation can be easily performed.

[0041] The parameter calculation means 5b calculates the frequency parameter f by applying the steepest descent method. m ',φ m After converging to a certain extent, it is desirable to apply the so-called Newton method to achieve high-precision convergence. Specifically, as the Newton method, the frequency parameter f m ',φ m ' is required. In addition, in the formula (7) and the formula (8), J is the formula (9). m is μ m Similarly, it is a weighting coefficient based on the deceleration method and takes values ​​from 0 to 1 as appropriate.

[0042]

number

[0043]

number

[0044]

number

[0045] The parameter calculation means 5b calculates a frequency parameter f by using the formulas (7) and (8). m ',φ m After calculating ', the frequency parameter A is calculated by the formula (6) in the same way as the steepest descent method. m ' is converged and this series of calculations is further iterated. In this way, the parameter calculation means 5b can estimate the frequency parameters A', f', and φ' quickly and accurately by using a solution that combines the steepest descent method and the Newton method.

[0046] In this way, the signal analysis means 5 receives the shaped signal and derives the frequency component with the maximum amplitude as a specific frequency indicating the distance to the target 14. Such a signal analysis method can be used in combination with a harmonic revealing process (see JP 2022-7811 A). The analysis result output means 5c outputs the analysis result to the display device 26, a printer, or other devices via an analysis result output unit (not shown).

[0047] The calculation means 6 derives the distance to the target and / or the movement speed from the specific frequency of the reflected wave, converting the specific frequency derived by the signal analysis means 5 into a distance to measure the distance to the target, and converting the specific frequency derived by the signal analysis means 5 into a speed to measure the movement speed of the object to be measured. The signal analysis means 5 uses a signal analysis method (NHA) that removes side lobes to make the main lobe apparent, so that the distance to the target 14 can be derived with high accuracy even if the measurement object is close by, depending on the frequency band of the chirp signal (see Figure 5).

[0048] When the FMCW radar of the above system is installed at positions 3.0 m and 1.0 m away from the target 14, which moves ±1 (mm), and the frequency of the beat signal is derived, the beat frequency f fm (Hz) is derived from equation (10), and the distance resolution Δx is derived from equation (11).

[0049]

number

[0050]

number

[0051] where f fm : beat frequency (Hz), W: chirp bandwidth (Hz), T: chirp time (s), c: speed of light (m / s), x: distance to target (m). W: chirp bandwidth 150 (MHz), T: chirp time 1 (ms), c: speed of light 3×10 8 (m / s), x: Beat frequency f when distance to target is 3 (m) and 1 (m) fm is given by equations (12) and (13), and the distance resolution Δx is given by equation (14).

[0052]

number

[0053]

number

[0054]

number

[0055] In this way, when FFT is adopted as the signal analysis method, the resolution is limited to 1.00 m as described above (see FIG. 8), and the frequency resolution is 1000 (Hz)×1.00 (m)=1000 Hz. On the other hand, the system according to the present disclosure having the above-mentioned configuration can detect a fluctuation range of ±100 (Hz) as shown in Fig. 9, and can perform signal analysis with approximately 10 times the accuracy of FFT. It was also confirmed that the system can accurately measure heart rate information even in the field of detecting human vital signs.

[0056] The calculation means 6 is equipped with a bio-information extraction means 6a that detects frequencies at which biological fluctuations exist over time in the distance or movement speed derived by the calculation means 6 as signals related to vital information and movement information of the living body (hereinafter, these will be referred to as "biological information") and derives specified bio-information of the living body that is the target 14, such as heart rate, breathing, and movement. For example, when the heart is at rest, it repeatedly contracts at a constant rhythm, but it is known that the heartbeat period shortens when excited. This is a fluctuation on a relatively long time scale that appears in the heartbeat, but the heartbeat period also varies on a relatively short time scale. In this way, biological phenomena have fluctuations on a scale that do not exist in the waves emitted by artificial objects. Here, such fluctuations are referred to as "biological fluctuations." Biological fluctuations are characterized by an increase or decrease of about 10 percent in the heartbeat period even under normal circumstances, and other biological information is also characterized by "1 / f fluctuations" or "1 / f 2 Fluctuations are said to exist. The bio-information extraction means 6a detects, as bio-information, the frequency derived by the signal analysis means 5 or the distance derived by the calculation means 6, where there is a fluctuation that matches the fluctuation width of each of the bio-information including the heart rate at rest or during exercise. The biometric information extraction means 6a provides the biometric information having biological fluctuations to a system for evaluating the biometric information. [Explanation of symbols]

[0057] 1 transmitting means, 2 receiving means, 3 data recording means, 4 signal shaping means, 4a distortion removal means, 4b waveform appropriation means, 4c distortion correction means, 5 signal analysis means, 5a data storage means, 5b parameter calculation means, 5c analysis result output means, 6 calculation means, 6a Biometric information extraction means, 10 Oscillation unit, 11 Transmission unit, 12 Reception unit, 13 Analysis unit, 14 Target, 20 CPU, 21 ROM, 22 RAM, 23 storage device, 24 input operation control device, 25 signal input device, 26 display device, 30 mixer, 31 distortion elimination filter, 32 power amplifier, 33 low noise filter, 34 transmitting antenna, 35 receiving antenna, 36 90 degree phase shift circuit,

Claims

1. A signal analysis method for measuring a distance or a motion speed of a measurement target in a radar system, comprising: A step of generating a chirp signal whose frequency periodically increases and decreases with a constant modulation width, and radiating the chirp signal into space as a traveling wave; detecting the traveling wave reflected by a measurement target as a reflected wave, and generating a beat signal from the reflected wave and the chirp signal; generating a shaped signal by removing distortion components from the beat signal generated by the receiving means; performing a signal analysis of the shaped signal to derive a frequency component having a maximum amplitude as a specific frequency reflected by the measurement target; Do the following: The signal analysis involves determining initial values ​​of the parameters so that a sum of squares of a difference between the shaped signal and a sinusoidal model signal expressed as parameters of a frequency f', a phase φ', and an amplitude A' is minimized, and converging from the determined initial values ​​to an optimal solution of a nonlinear equation, The shaped signal is a one-dimensional measurement s(x ｎ ) and the frequency f is a frequency at which the right side of the following formula (1) becomes a minimum value, where N is the number of data. ｘ The signal analysis method is characterized by obtaining a signal φ′, a phase φ′ and an amplitude A′. [0010]

2. A signal analysis method for measuring a distance or a motion speed of a measurement target in a radar system, comprising: A step of generating a chirp signal whose frequency periodically increases and decreases with a constant modulation width, and radiating the chirp signal into space as a traveling wave; detecting the traveling wave reflected by a measurement target as a reflected wave, and generating a beat signal from the reflected wave and the chirp signal; generating a shaped signal by removing distortion components from the beat signal generated by the receiving means; performing a signal analysis of the shaped signal to derive a frequency component having a maximum amplitude as a specific frequency reflected by the measurement target; Do the following: The signal analysis involves determining initial values ​​of the parameters so that a sum of squares of a difference between the shaped signal and a sinusoidal model signal expressed as parameters of a frequency f', a phase φ', and an amplitude A' is minimized, and converging from the determined initial values ​​to an optimal solution of a nonlinear equation, Among the parameters, a frequency f' and a phase φ' are converged by applying the steepest descent method, and then a Newton method incorporating a deceleration method is applied to the parameters, and then an amplitude A' is converged by applying the steepest descent method, and then a Newton method incorporating a deceleration method is applied to the parameters, and then the amplitude A' is converged by applying the steepest descent method, and then the Newton method incorporating a deceleration method is applied to the parameters, The shaped signal is a one-dimensional measurement s(x ｎ ) and the frequency f is a frequency at which the right side of the following formula (1) becomes a minimum value, where N is the number of data. ｘ The signal analysis method is characterized by obtaining a signal φ′, a phase φ′ and an amplitude A′. [0025]

3. A radar system for measuring a distance and / or a speed of movement of a measurement object, comprising: A transmitting means for generating a chirp signal whose frequency periodically increases and decreases with a constant modulation width and radiating the chirp signal into space as a traveling wave; a receiving means for detecting the traveling wave reflected by a measurement target as a reflected wave and generating a beat signal from the reflected wave and the chirp signal; a signal shaping means for generating a shaped signal by removing distortion components from the beat signal generated by the receiving means; A signal analysis means for deriving a frequency component having a maximum amplitude as a specific frequency reflected by the measurement target based on a signal analysis of the shaped signal; A radar system comprising:

4. the signal analysis means determines initial values ​​of the parameters so that a sum of squares of a difference between the shaped signal and a sinusoidal model signal expressed as parameters of a frequency f', a phase φ', and an amplitude A' becomes a minimum value, and causes the signal analysis means to converge to an optimal solution of a nonlinear equation from the determined initial values, The shaped signal is a one-dimensional measurement s(x ｎ ) and the frequency f is a frequency at which the right side of the following formula (1) becomes a minimum value, where N is the number of data. ｘ 4. The radar system according to claim 3, wherein the signal φ′, phase φ′ and amplitude A′ are calculated. [0030]

5. the signal analysis means determines initial values ​​of the parameters so that a sum of squares of a difference between the shaped signal and a sinusoidal model signal expressed as parameters of a frequency f', a phase φ', and an amplitude A' becomes a minimum value, and causes the signal analysis means to converge to an optimal solution of a nonlinear equation from the determined initial values, Among the parameters, a frequency f' and a phase φ' are converged by applying the steepest descent method, and then a Newton method incorporating a deceleration method is applied to the parameters, and then an amplitude A' is converged by applying the steepest descent method, and then a Newton method incorporating a deceleration method is applied to the parameters, and then the amplitude A' is converged by applying the steepest descent method, and then the Newton method incorporating a deceleration method is applied to the parameters, The shaped signal is a one-dimensional measurement s(x ｎ ) and the frequency f is a frequency at which the right side of the following formula (1) becomes a minimum value, where N is the number of data. ｘ 4. The radar system according to claim 3, wherein the signal φ′, phase φ′ and amplitude A′ are calculated. [0045]

6. 4. The radar system according to claim 3, wherein the signal shaping means generates the shaped signal by linearization processing using a non-negative least squares method for a chirp time period in the beat signal.

7. The signal shaping means comprises: a distortion removing means for performing a distortion removing process to remove a distortion region in a chirp time period in the beat signal as an invalid analysis section; a distortion correcting means for correcting the remaining distortion of the beat signal after the invalid analysis section has been removed, and / or a waveform applying means for applying a waveform approximating an ideal waveform with suppressed distortion to a missing portion resulting from the distortion removal process to generate a shaped signal; 4. The radar system of claim 3, further comprising:

8. 4. The radar system according to claim 3, wherein the transmitting means generates a chirp signal having a triangular or sawtooth waveform that periodically increases and decreases in a certain frequency band, and radiates the chirp signal into space as a traveling wave.

9. a calculation means for calculating a distance from the transmitting means to a measurement object from the specific frequency derived by the signal analysis means, 4. The radar system according to claim 3, wherein the frequency derived by said signal analysis means or the distance derived by said calculation means is detected as biological information when biological fluctuations exist therein.

Images