Driver safety monitoring system and method based on HRV autonomic nervous function parameters

Abstract

The present invention discloses a driver safety monitoring system and method based on HRV autonomic nerve function parameters, comprising the following steps: using a sliding collar type ECG electrode arranged on a steering wheel to collect the driver's ECG signal, transmitting the collected ECG signal to an ECG amplifier for amplification, converting it into an ECG digital signal after amplification, and saving it; performing time domain analysis and calculation on the RR interval and the RR interval variation in the ECG digital signal, obtaining the time domain parameters of average heart rate, standard deviation SDNN, root mean square RMSSD of continuous RR intervals, and percentage PNN50 of continuous RR intervals with a difference of more than 50ms; performing frequency domain analysis and calculation on the RR interval in the ECG digital signal, obtaining the power value and spectrum envelope diagram of the autonomic nerve function frequency domain parameter LF / HF; judging whether the driver's state is abnormal according to the power value and spectrum envelope diagram of the time domain parameters and frequency domain parameters. The present invention ensures the accuracy of early warning and improves the sensitivity of safety monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of electrocardiogram signal detection and analysis, and in particular to a driver safety monitoring system and method based on HRV autonomic nerve function parameters. Background Art

[0002] On a highway with a high speed of 80 kilometers per hour, the clock "ticks" in one second, and the car travels more than 20 meters (the speed of a train on a railroad is even faster, reaching more than 200 kilometers per hour). If the driver falls asleep due to excessive fatigue at this moment, the steering wheel will lose control and a traffic accident will easily occur, causing heavy losses or even death. Drivers are exposed to a variety of occupational hazards such as noise, vibration, gasoline, and CO when driving. Long-term exposure can cause emotional changes such as depression, anxiety, irritability, fatigue, and distraction. The whole body vibration of the driver while driving can also reduce the speed of nerve conduction in the body, resulting in reduced operating ability. Practice shows that after driving for 3 hours continuously, the driver's alertness level will become poor, and the body's cognitive state and physiological indicators will change. If the driver can receive an alarm in time at this time, he can improve his own alertness, cognitive state and driving behavior performance to reduce traffic safety risks. The accurate physiological measurement of HRV autonomic nervous function parameters is generally obtained by contacting (i.e. placing metal electrode sensors on relevant parts of the human body) the RR interval signal of the heart beat. Figure 1 A series of continuous RR intervals or instantaneous heart rates can be used to record changes in heart rate over time, so it is also called a heart rate variability (HRV) signal. The ECG of a healthy person shows obvious periodicity, but careful observation and measurement of the ECG show that the ECG of a normal person also fluctuates slowly and slightly within a certain range. Two adjacent RR intervals (cardiac cycles) usually differ by tens of milliseconds or even more than a hundred milliseconds. Heart rate variability refers to the slight changes in instantaneous heart rate over time, usually defined as the slight fluctuations in the RR intervals or instantaneous heart rate of a continuous normal heartbeat.

[0003] The prior art (CN117113053A) discloses a method of obtaining the heart rate variability analysis of the heartbeat signal of the target driver based on FMCW millimeter wave radar to obtain physiological characteristics. Human body radar is a technology that detects human activities through radio waves or microwaves. It can monitor physiological activities such as human breathing and heartbeat, and then convert this information into signals for further analysis and processing. Human body radar detects surrounding objects, including the human body, by emitting electromagnetic waves and using the reflection and scattering of electromagnetic waves by objects. When electromagnetic waves interact with the human body, they are absorbed, reflected and scattered. This interaction changes the phase and amplitude of the electromagnetic waves, and the receiver can capture and analyze these changes. The heartbeat is generated by the contraction and relaxation process of the heart, which is caused by the activity of the heart muscle. When blood is pushed by the heart, the heart fills and contracts, generating a heartbeat signal. This process causes tiny movements and expansions of the chest, changing the size of the chest, and the human body radar can detect this tiny change; by analyzing these signals, information such as the frequency and intensity of the heartbeat can be extracted. However, the small changes in chest size (phase and amplitude) caused by the reflection and scattering of the heartbeat are easily affected by various factors (including various electromagnetic interference noises in space, the shaking of the human body, the breathing of the human body, the thickness of the clothes worn by the human body, etc.), which makes the detected RR interval signal reflecting the human heart beat seriously distorted and inaccurate, thus losing its application value. There is also a non-contact method of measuring heartbeat using camera customized software in the prior art. The result of this heart rate detection is also inaccurate. If the measurement continues for a period of time, the heartbeat will stabilize at a value.

[0004] Therefore, it is urgent to solve the above problems. Summary of the invention

[0005] Purpose of the invention: The first purpose of the present invention is to provide a driver safety monitoring method based on HRV autonomic nervous function parameters that can accurately monitor the driver's electrocardiogram signals and provide early warning.

[0006] The second object of the present invention is to provide a driver safety monitoring system based on HRV autonomic nervous function parameters.

[0007] Technical solution: To achieve the above purpose, the present invention discloses a driver safety monitoring method based on HRV autonomic nerve function parameters, comprising the following steps:

[0008] (1) using a sliding ring-type ECG electrode disposed on a steering wheel to collect the driver's ECG signal, the sliding ring-type ECG electrode comprising three ring-type electrode sheets disposed on the steering wheel and slidable, with an insulating isolation zone disposed between adjacent ring-type electrode sheets;

[0009] (2) transmitting the collected ECG signals to an ECG amplifier for amplification, converting them into ECG digital signals after amplification and saving them;

[0010] (3) The RR intervals and RR interval variations in the ECG digital signal were analyzed and calculated in the time domain to obtain the time domain parameters of mean heart rate HR, standard deviation SDNN, root mean square RMSSD of consecutive RR intervals, and percentage PNN50 of consecutive RR intervals with a difference of more than 50 ms;

[0011] (4) Perform frequency domain analysis and calculation on the RR interval and RR interval variation in the ECG digital signal, calculate the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method, and obtain the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF of the ECG data to be analyzed by the AR model method;

[0012] (5) If the time domain parameter average heart rate If more than two of the standard deviation SDNN, the root mean square RMSSD of consecutive RR intervals, the percentage PNN50 of consecutive RR intervals with a difference of more than 50ms, and the power values ​​of the frequency domain parameters LF / HF are not within the normal threshold range, it indicates that the driver's state is abnormal; or if the spectrum envelope diagram of the ECG data to be analyzed is within the standard group diagram, it indicates that the driver's state is abnormal.

[0013] Among them, the average heart rate in step (3) The calculation formula is:

[0014]

[0015] Where RR i It represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms; N is the length of the RR interval sequence to be analyzed.

[0016] Preferably, the calculation formula for the standard deviation SDNN of the RR interval in step (3) is:

[0017]

[0018] Where N is the length of the RR interval sequence to be analyzed, RR i Represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms. is the average value of the RR interval of N heart beats.

[0019] Furthermore, the calculation formula for the root mean square RMSSD of consecutive RR intervals in step (3) is:

[0020]

[0021] Where N is the length of the RR interval sequence to be analyzed, RR iRepresents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms; RR i+1 It represents the value of the i+1th RR interval in the RR interval sequence to be analyzed, in ms, and the unit of the root mean square RMSSD is ms.

[0022] Furthermore, the calculation formula for the percentage of consecutive RR intervals with a difference of more than 50 ms in step (3) is:

[0023] PNN50 = (NN50 / total number of RR intervals) x 100%

[0024] NN50 is the number of heart beats in which the difference between adjacent RR intervals among all RR intervals is greater than 50 ms, in units of beats.

[0025] Preferably, the specific steps of calculating the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method in step (4) are:

[0026] The power spectral density PSD is calculated by performing fast Fourier transform on the RR interval data, where the power spectral density PSD can be expressed as a function of frequency, denoted as P(f), as follows:

[0027]

[0028] In the formula, x(n) (n = 0, 1, ..., N-1) represents the RR interval sequence of the heart beat, N represents the length of the RR interval sequence to be analyzed, usually N = 256 or N = 512; Δt is the average sampling interval of the RR interval sequence, that is, the average value of the RR interval sequence in this section X(f) is the discrete Fourier transform of x(n);

[0029] The power spectral density PSD is divided into the following four frequency bands: Ultra-low frequency band ULF: <0.003HZ, which is PSD ULF , Very low frequency band VLF: 0.003~0.04HZ, which is PSD VLF , low frequency band LF: 0.04~0.15HZ, which is PSD LF , high frequency band HF0.15~0.40HZ, that is PSD HF , total frequency band: ≤0.4HZ, which is PSD 总 ;

[0030] Heart rate variability power spectrum PSD of the total frequency band 总 =PSD ULF +PSD LF +PSD HF +PSD VLF ,

[0031] LFnorm=100×PSD LF / (PSD 总 -PSD VLF ),

[0032] HFnorm = 100 × PSD HF / (PSD 总 -PSD VLF ),

[0033] LF / HF=PSD LF / PSD HF .

[0034] Furthermore, the specific steps of obtaining the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the AR model method in step (4) are as follows:

[0035] The difference equation of the AR model, i.e. the autoregressive model, can be expressed as:

[0036]

[0037] Where x(n) is the RR interval sequence of the heart beat, and u(n) is a sequence with a mean of zero and a variance of σ 2 white noise sequence; p is the model order, p = 15, a k is the coefficient of the AR model, k = 1, 2, 3, …, p;

[0038] According to the autocorrelation function of x(n), the canonical equation of the AR model is constructed, namely the Yule-Walker equation. After solving it, the coefficients a are obtained. k and σ 2 The estimated value of x(n) is AR (f) can be expressed as:

[0039]

[0040] The envelope diagram of the ECG data to be analyzed, which is a power density spectrum curve of the RR interval every 5 minutes, is obtained, that is, the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF.

[0041] Furthermore, the method for selecting the determined value of the model order p in step (4) is as follows: performing frequency domain analysis and calculation on the RR interval and the RR interval variation in the ECG digital signal, calculating the power values ​​of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function by the average period method, and drawing the corresponding power density spectrum curve; setting the initial value of the model order p, and obtaining the envelope diagram of the power density spectrum curve of the RR interval by the AR model method, that is, the spectrum diagram of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function; if the power density spectrum curve matches the spectrum envelope diagram, then the model order p is the determined value; if the power density spectrum curve does not match the spectrum envelope diagram, then adjusting the value of the model order p, and re-obtaining the envelope diagram of the power density spectrum curve of the RR interval by the AR model method, until the power density spectrum curve matches the spectrum envelope diagram, and obtaining the determined value of the model order p.

[0042] Preferably, the method for selecting the standard group graph in step (5) is: selecting the ECG data from the process of being awake to the state of drowsiness as the standard ECG data, inputting the standard ECG data into the AR model to obtain a cluster of standard AR model graphs, and then selecting several graphs with a similarity greater than 0.7 to the standard AR model graphs to form a standard group graph.

[0043] The present invention discloses a driver safety monitoring system based on HRV autonomic nerve function parameters, comprising a sliding collar type ECG electrode, the sliding collar type ECG electrode is arranged on a steering wheel, the sliding collar type ECG electrode collects the driver's ECG signal, the sliding collar type ECG electrode comprises three collar type electrode sheets which are sleeved on the steering wheel and can slide, and an insulating isolation belt is arranged between adjacent collar type electrode sheets;

[0044] The signal conversion and storage module transmits the collected ECG signals to the ECG amplifier for amplification, converts them into ECG digital signals after amplification, and saves them;

[0045] The time domain calculation module is used to perform time domain analysis and calculation on the RR interval and RR interval variation in the ECG digital signal to obtain the average heart rate of the time domain parameter The standard deviation SDNN, the root mean square RMSSD of consecutive RR intervals, and the percentage of consecutive RR intervals that differ by more than 50 ms PNN50;

[0046] The frequency domain calculation module is used to perform frequency domain analysis and calculation on the RR interval and RR interval variation in the ECG digital signal, calculate the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method, and obtain the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF of the ECG data to be analyzed by the AR model method;

[0047] Abnormal judgment module, in which if the time domain parameter average heart rate If more than two of the standard deviation SDNN, the root mean square RMSSD of consecutive RR intervals, the percentage PNN50 of consecutive RR intervals with a difference of more than 50ms, and the power values ​​of the frequency domain parameters LF / HF are not within the normal threshold range, it indicates that the driver's state is abnormal; or if the spectrum envelope diagram of the ECG data to be analyzed is within the standard group diagram, it indicates that the driver's state is abnormal.

[0048] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the present invention collects the driver's electrocardiogram signal, performs time domain and frequency domain analysis on the electrocardiogram signal, obtains the time domain parameter value and the power value and spectrum envelope diagram of the frequency domain parameter respectively, and then judges whether the driver's state is abnormal based on the time-frequency parameter analysis, thereby ensuring the accuracy of the early warning and improving the sensitivity of safety monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of ECG waveform in the present invention;

[0050] Figure 2 It is a schematic diagram of the process of the present invention;

[0051] Figure 3 This is a schematic diagram of the FFT power spectrum of the HRV signal in a 5-minute period in the present invention;

[0052] Figure 4 It is a schematic diagram of the AR model curve and FFT power spectrum of the HRV signal in a 5-minute period in the present invention;

[0053] Figure 5 It is a schematic diagram of the AR model curve of the HRV signal in a 5-minute period in the present invention;

[0054] Figure 6 It is a graphical diagram of PSD in the process from wakefulness to sleepiness in the simulation experiment of the present invention;

[0055] Figure 7 It is a schematic diagram of the flow of time-frequency analysis in the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0057] Embodiment 1:

[0058] The human heart rate (measured in beats per minute or bpm) is the number of heartbeats in 1 minute. Electrocardiogram (ECG) is a physiological method of recording the electrical activity of the heart. The individual heartbeat waves in the ECG signal show changes in electrical potential, and its components are called P waves, QRS complexes, and T waves, such as Figure 1 shown.

[0059] The autonomic nervous system is mainly composed of the sympathetic and parasympathetic nervous systems. The autonomy of the heart is an inherent characteristic of the cardiac pacemaker tissue, but the heart rate and rhythm are largely regulated by the autonomic nervous system. The heart rate is not absolutely regular. There is a time difference of tens of milliseconds or even more than one hundred milliseconds between two heartbeats. The change of the heartbeat interval of a healthy person is due to the change of the sympathetic and parasympathetic nerves with factors such as breathing, which is a normal phenomenon. The time difference between successive heartbeats, that is, the slight fluctuation of the R-R interval of successive heartbeats, is usually defined as heart rate variability HRV. Therefore, heart rate variability HRV can reflect the activity of the autonomic nerves. The heart rate variation signal in the present invention adopts the electrocardiogram signal of the RR interval sequence. After a long journey, the driver often eventually causes fatigue of the body's nervous system due to various reasons. From the perspective of human physiology, fatigue is a protective reaction issued by the body. The appearance of fatigue, from the perspective of the autonomic nervous system, refers to the increased activity of the parasympathetic nerve (also known as the vagus nerve) system. When the driver is tired, the time domain analysis of the ECG heart rate variability HRV signal shows that his heart rate HR will decrease; the standard deviation SDRR of the difference between the R-R intervals of successive heart beats will become smaller; and the frequency domain analysis of the ECG heart rate variability HRV signal shows that the power of its high-frequency component HF0.15-0.4Hz will increase. At this time, the driver should be reminded to slow down, stop driving and take a rest to prevent traffic accidents. Specifically, three slidable ring-type electrodes with red, yellow and blue markings can be set on the steering wheel in the driver's studio of a car or train. The three ring-type electrodes should be connected, and there is an insulating isolation zone of about 3mm in the middle. The three ring-type electrodes should be connected to the input end of the single-lead ECG signal collector according to engineering requirements; the driver's right hand pinches the red ring on the steering control wheel, and the left hand pinches the yellow and blue connected ring. The body's MV-level single-lead ECG signal is guided to the input end of the ECG amplifier through these three ring lead electrodes for amplification. After being amplified several hundred times, it is sampled and converted into a digital signal by the A / D analog-to-digital conversion module with a sampling rate of 1kHz, and the ECG digital signal is stored in the storage module. The data is read and the ECG data (RR interval and RR interval variation) is analyzed and calculated in the time domain and frequency domain (low frequency 0.04-0.15Hz power LF and high frequency 0.15-0.4Hz power HF) to obtain the time domain parameter average heart rate Standard deviation SDNN, root mean square RMSSD of consecutive RR intervals, percentage PNN50 of consecutive RR intervals with a difference of more than 50ms, and frequency domain parameters of autonomic nervous function LFnorm, HFnorm, LF / HF. If several of these seven parameters exceed the threshold, the alarm system of the safety monitoring equipment will sound an alarm to wake the driver to return to normal physiological state, relieve fatigue, or stop driving to rest to prevent traffic accidents.

[0060] The ECG of a healthy person shows obvious periodicity. However, careful observation and measurement of the ECG show that the ECG of a normal person also fluctuates slowly and slightly within a certain range. Two adjacent RR intervals (cardiac cycles) usually differ by tens of milliseconds or even more than a hundred milliseconds. Heart rate variability (HRV) refers to the slight changes in instantaneous heart rate over time, usually defined as the slight fluctuations in the RR intervals or instantaneous heart rate of consecutive normal heartbeats. Figure 1 A series of continuous RR intervals or sequences of instantaneous heart rates can be used to record changes in heart rate over time, and are therefore also called heart rate variability signals. The heart rate variability signal of the present invention uses an RR interval sequence electrocardiographic signal. The autonomy of the heart is an inherent characteristic of the cardiac pacemaker tissue, but the heart rate and rhythm are largely regulated by the ANS (Autonomic Nerve System). The PPS (Parasympathetic System) affects the heart rate by releasing acetylcholine; the SPS (Sympathetic System) affects the heart rate by releasing adrenaline and noradrenaline. Heart rate variability parameters are divided into time domain parameters and frequency domain parameters. HRV is the variability of the time interval between adjacent heartbeats. HRV can be derived from ECG data, and the time range is set to 1 to 5 minutes. Heart rate can be used as an indicator of physiological arousal that reflects changes in driving demand. Compared with performing a single driving task in a simulator, when performing a dual visual and auditory task, the heart rate (HR) will increase with the increase in workload due to the increased cognitive demand. HR and HRV are also sensitive to low arousal states, such as alertness and sleepiness. During a 3-hour continuous driving task, the indicator of decreased alertness was a significant decrease in heart rate over time. Sleepiness in car drivers and airplane pilots may also be associated with a decrease in HR. HRV indicators can be roughly divided into the time domain and frequency domain of HRV. Based on the time domain parameters, there are changes in heart rate HR and heart beat intervals, such as the instantaneous heart rate (i.e., instantaneous RR interval), average heart rate, and frequency domain of HRV. The parameters include the standard deviation SDNN of the RR interval, the percentage PNN50 of consecutive RR intervals with a difference of more than 50s, and the root mean square RMSSD of consecutive RR intervals.

[0061] The present invention discloses a driver safety monitoring method based on HRV autonomic nerve function parameters, comprising the following steps:

[0062] (1) using a sliding ring-type ECG electrode disposed on a steering wheel to collect the driver's ECG signal, the sliding ring-type ECG electrode comprising three ring-type electrode sheets disposed on the steering wheel and slidable, with an insulating isolation zone disposed between adjacent ring-type electrode sheets;

[0063] Three red, yellow and blue sliding ring-type electrodes are set on the steering wheel in the car or train driver's studio. The three ring-type electrodes should be connected, with an insulating isolation zone of about 3mm in the middle. The electrode on the right half of the steering wheel is red, and the two electrodes on the left half of the steering wheel are yellow and blue. The three ring-type electrodes should be connected to the input end of the single-lead ECG signal collector according to engineering requirements, such as Figure 2 As shown;

[0064] (2) transmitting the collected ECG signals to an ECG amplifier for amplification, converting them into ECG digital signals after amplification, and saving them;

[0065] (3) Perform time domain analysis and calculation on the RR interval and RR interval variation in the ECG digital signal to obtain the average heart rate of the time domain parameter The standard deviation SDNN, the root mean square RMSSD of consecutive RR intervals, and the percentage of consecutive RR intervals that differ by more than 50 ms PNN50;

[0066] Average heart rate The calculation formula is:

[0067]

[0068] Where RR i represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms; N is the length of the RR interval sequence to be analyzed;

[0069] The calculation formula for the standard deviation SDNN of the RR interval is:

[0070]

[0071] Where N is the length of the RR interval sequence to be analyzed, RR i Represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms. is the average RR interval of N heart beats;

[0072] The formula for calculating the root mean square RMSSD of consecutive RR intervals is:

[0073]

[0074] Where N is the length of the RR interval sequence to be analyzed, RR i Represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms; RR i+1 It represents the value of the i+1th RR interval in the RR interval sequence to be analyzed, in ms, and the unit of the root mean square RMSSD is ms;

[0075] The percentage of consecutive RR intervals that differ by more than 50 ms, PNN50, is calculated as:

[0076] PNN50 = (NN50 / total number of RR intervals) x 100%

[0077] NN50 is the number of heart beats in which the difference between adjacent RR intervals is greater than 50ms among all RR intervals, in units;

[0078] (4) Perform frequency domain analysis and calculation on the RR interval and RR interval variation in the ECG digital signal to obtain the frequency domain parameters of autonomic nervous function LFnorm, HFnorm, and LF / HF, such as Figure 7 As shown;

[0079] The power values ​​of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function were calculated by the average cycle method, and the spectrum envelope diagrams of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function were obtained by the AR model method.

[0080] The specific steps for calculating the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method are as follows:

[0081] The power spectral density PSD is calculated by performing fast Fourier transform on the RR interval data, where the power spectral density PSD can be expressed as a function of frequency, denoted as P(f), as follows:

[0082]

[0083] In the formula, x(n) (n = 0, 1, ..., N-1) represents the RR interval sequence of the heart beat, N represents the length of the RR interval sequence to be analyzed, usually N = 256 or N = 512; Δt is the average sampling interval of the RR interval sequence, that is, the average value of the RR interval sequence in this section X(f) is the discrete Fourier transform of x(n);

[0084] The power spectral density PSD is divided into the following four frequency bands: Ultra-low frequency band ULF: <0.003HZ, which is PSD ULF , Very low frequency band VLF: 0.003~0.04HZ, which is PSD VLF , low frequency band LF: 0.04~0.15HZ, which is PSD LF , high frequency band HF0.15~0.40HZ, that is PSD HF , total frequency band: ≤0.4HZ, which is PSD 总The high-frequency band HF component in the heart rate variability signal mainly reflects the activity of the cardiac vagus nerve, and the low-frequency band LF component reflects the cardiac sympathetic nerve activity or the sympathetic vagus joint activity. Therefore, the ratio of the low-frequency band component of the signal LF / HF reflects the balance of sympathetic and vagal activities.

[0085] Heart rate variability power spectrum PSD of the total frequency band 总 =PSD ULF +PSD LF +PSD HF +PSD VLF ,

[0086] LFnorm=100×PSD LF / (PSD 总 -PSD VLF ),

[0087] HFnorm = 100 × PSD HF / (PSD 总 -PSD VLF ),

[0088] LF / HF=PSD LF / PSD HF ,

[0089] The specific steps of obtaining the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the AR model method are as follows:

[0090] The difference equation of the AR model, i.e. the autoregressive model, can be expressed as:

[0091]

[0092] Where x(n) is the RR interval sequence of the heart beat, and u(n) is a sequence with a mean of zero and a variance of σ 2 white noise sequence; p is the model order, p = 15, a k is the coefficient of the AR model, k = 1, 2, 3, …, p;

[0093] According to the autocorrelation function of x(n), the canonical equation of the AR model is constructed, namely the Yule-Walker equation. After solving it, the coefficients a are obtained. k and σ 2 The estimated value of x(n) is AR (f) can be expressed as:

[0094]

[0095] Obtaining the envelope diagram of the ECG data to be analyzed, which is a power density spectrum curve of the RR interval every 5 minutes, is the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF;

[0096] The method for selecting the determined value of the model order p is as follows: the RR interval and the RR interval variation in the ECG digital signal are analyzed and calculated in the frequency domain, the power values ​​of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function are calculated by the average period method, and the corresponding power density spectrum curve is drawn; the initial value of the model order p is set, and the envelope diagram of the power density spectrum curve of the RR interval is obtained by the AR model method, that is, the spectrum diagram of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function; if the power density spectrum curve matches the spectrum envelope diagram, the model order p is the determined value; if the power density spectrum curve does not match the spectrum envelope diagram, the value of the model order p is adjusted, and the envelope diagram of the power density spectrum curve of the RR interval is obtained again by the AR model method until the power density spectrum curve matches the spectrum envelope diagram, and the determined value of the model order p is obtained;

[0097] Frequency domain analysis is to convert the heart rate variability signal in the time domain into the frequency domain and find the power spectrum. There are two methods to find the power spectrum: the average period method and the AR model method. The average period method uses the fast Fourier transform FFT to find the power value, and the AR model method is used to obtain a smooth power spectrum line close to FFT. The average period method is based on FFT. It makes an assumption about the period extension of the signal. To obtain a higher spectral resolution, longer data is required. Because the heart rate variability signal records the R-R interval of successive heartbeats, it is a non-uniform sampling. In the actual FFT processing, a relatively long period of successive R-R intervals is often regarded as a uniform time series, and the sampling rate is defined as is the average RR interval of this data. If the collected RR interval of heart rate data is transformed by FFT with a step length of 5 minutes, the FFT power spectrum of the HRV signal frequency domain in this 5-minute period can be obtained, as shown in Figure 3 shown.

[0098] like Figure 4 and Figure 5 As shown in the figure, the AR model method establishes a parameter model for the heart rate signal and reflects the characteristics of the signal by transferring the parameters of the function. The AR model method belongs to the modern spectrum estimation method, which requires short data and high resolution. Compared with the spectrum of FFT analysis, the AR model curve is like a smooth envelope, such as Figure 4 The red smooth curve and Figure 5 As the order of the AR model increases, the curve of the AR model fluctuates more and more, and is closer to the spectrum line of FFT, such as Figure 4 This is shown in the blue curve. This shows that the results of the two analysis methods are very consistent, so for the frequency domain analysis of non-stationary heart rate variability signals, the AR model is often used to draw the spectrum line of the analysis, and the FFT method is used to calculate the power spectrum density value.

[0099] The frequency domain analysis method of the autonomic nervous function of the heart rate variability HRV signal is based on the two frequency bands of low frequency band LF (0.04-0.15Hz) and high frequency band HF (0.15-0.4Hz). The change of the driver's sleepiness will affect the frequency domain parameters of the autonomic nervous function of the HRV signal, such as the power spectrum LF-HRV and HF-HRV changes with fatigue and sleepiness, such as Figure 6 As shown in the figure, a simulation experiment was conducted on a young 22-year-old man, from wakefulness to sleepiness. In the autonomic nervous system regulation mechanism, the energy changes of low and medium frequency peaks are affected by both the sympathetic nerve and the vagus nerve, while the high frequency peak is only affected by the vagus nerve. In clinical practice, 0.15Hz is often used as the boundary to divide it into low-frequency components (LF-HRV) and high-frequency components (HF-HRV). The LF-HRV component is affected by both the sympathetic nerve and the vagus nerve, among which the sympathetic nerve is dominant. The HF-HRV component reflects the activity of the vagus nerve. Therefore, the HF peak of a healthy person will increase when he is asleep, decrease when he is awake, and even lower when he is excited. Figure 6 The peaks of the HF in the first 19 curves are very low, and the last one suddenly becomes high because the subject fell asleep at the end of the experiment. When the subject is anxious or excited, the high-frequency peak should decrease, but sleep will offset this decrease, so the subject is required to stay awake and quiet. Therefore, analysis based on the frequency domain of HRV can be used to detect fatigue status earlier and sensitively, issue warnings, and improve attention and behavioral performance.

[0100] (5) If the time domain parameter average heart rate If more than two of the standard deviation SDNN, the root mean square RMSSD of consecutive RR intervals, the percentage PNN50 of consecutive RR intervals with a difference of more than 50ms, and the frequency domain parameters LF / HF are not within the normal threshold range, it indicates that the driver's state is abnormal; if the envelope diagram of the ECG data to be analyzed is within the standard group diagram, it indicates that the driver's state is abnormal; if the envelope diagram of the ECG data to be analyzed is not within the standard group diagram, it indicates that the driver's state is normal; when the driver's state is abnormal, an alarm will immediately sound in the car to wake him up to ensure safety and eliminate danger; the alarm level can also be divided into three levels: mild, moderate and severe according to the driver's fatigue level.

[0101] The specific method for selecting the standard group diagram is as follows:

[0102] The ECG data from wakefulness to drowsiness are selected as standard ECG data, and the standard ECG data are input into the AR model to obtain a cluster of standard AR model graphs. Then, several graphs with a similarity greater than 0.7 to the standard AR model graphs are selected to form a standard group graph.

[0103] The index thresholds recommended for time-frequency domain analysis parameters are shown in Tables 1 and 2.

[0104] Table 1 Threshold values ​​of short-term (5 min) time domain analysis parameters of autonomic nervous function

[0105]

[0106]

[0107] Table 2 Threshold values ​​of frequency domain analysis parameters of autonomic nervous function in short term (5 min)

[0108]

[0109] The above thresholds can be changed as the amount of data actually measured increases, and their threshold ranges can be changed through statistical analysis to make subsequent measurements more accurate.

[0110] The AR model method belongs to the modern spectrum estimation method. It uses the parameter model to separate and organically combine the randomness of the random process and a certain degree of predictability. The white noise excitation reflects the randomness of the process, and the deterministic model reflects the predictability of the process. Therefore, it can better summarize the properties of random signals. There are many advantages to using the parameter model method to study random signals. The power spectrum it obtains is a continuous function of frequency, which avoids the random fluctuations of the periodogram spectrum estimation, and a better spectrum estimation can be obtained with relatively short data, which is very beneficial for signals with strong non-stationarity such as heart rate variability.

[0111] Compared with the spectrum of FFT analysis, the AR model curve is like a smooth envelope, and as the AR model order increases, the AR model curve fluctuates more and is closer to the FFT spectrum. This shows that the results of the two analysis methods are very consistent. Because the heart rate variability signal is a relatively non-stationary data, it is more appropriate to use the AR model to analyze the spectrum.

[0112] There is a problem of determining the order when using the AR model to calculate the spectrum. After research, it is clear that the 15th order is most suitable for heart rate variability analysis. If the order is too high, the peak line will split and false spectrum peaks will appear; if the order is too low, the spectrum line cannot correctly reflect the changes in the spectrum peak. The order can be adjusted for different data. There are three algorithms for the AR model: L-D algorithm, Burg algorithm and Marple algorithm. Among them, the Marple algorithm is the fastest and has the smallest error, but the results of the three algorithms for heart rate variability analysis are basically the same.

[0113] The power spectrum of human heart rate variability generally ranges from 0 to 0.5 Hz, and it can have three spectral peaks:

[0114] (1) The first peak is around 0.03 Hz (0.02-0.09), which is a low-frequency peak and is related to multiple factors such as peripheral vasoconstriction, temperature regulation, and renin-angiotensin system activity;

[0115] (2) The second peak is around 0.10 Hz (0.09-0.15), which is a mid-frequency peak and is related to blood pressure regulation;

[0116] (3) The third peak is around 0.25 Hz (0.15-0.40), which is a high-frequency peak and is related to the heart rate changes caused by the respiratory cycle;

[0117] In the autonomic nervous system regulation mechanism, the energy changes of low and medium frequency peaks are affected by both the sympathetic nerves and the vagus nerves, while the high frequency peaks are only affected by the vagus nerves. In actual clinical applications, 0.15Hz is often used as the boundary to divide it into low-frequency component LF and high-frequency component HF. The LF component is affected by both the sympathetic nerves and the vagus nerves, with the sympathetic nerves being dominant. The HF component reflects the activity of the vagus nerve. Therefore, the HF peak of a healthy person will increase when he is asleep, decrease when he is awake, and even lower when he is excited. Figure 6 As shown in the figure, the peaks of the first 19 curves HF are very low, and the last one suddenly becomes high because the subject fell asleep at the end of the experiment. When the subject is anxious and excited, the high-frequency peak should decrease, but sleep will offset this decrease, so the subject is required to stay awake and quiet.

[0118] Embodiment 2: Embodiment 2 discloses a driver safety monitoring system based on HRV autonomic nervous function parameters, including a sliding collar type ECG electrode, which is arranged on a steering wheel, and the sliding collar type ECG electrode collects the driver's ECG signal. The sliding collar type ECG electrode includes three collar type electrode sheets that are slidable and sleeved on the steering wheel, and an insulating isolation zone is arranged between adjacent collar type electrode sheets; three red, yellow and blue slidable collar type electrode sheets are arranged on the steering control wheel in the driver's studio of a car or train, and the three collar type electrode sheets should be connected, and there is an insulating isolation zone of about 3 mm in the middle, respectively. The electrode sleeved on the right half of the steering wheel is red, and the two sleeved on the left half of the steering wheel are yellow and blue electrodes. The three collar type electrode sheets should be connected to the input end of the single-lead ECG signal collector according to engineering requirements, such as Figure 2 As shown;

[0119] The signal conversion and storage module transmits the collected ECG signals to the ECG amplifier for amplification, converts them into ECG digital signals after amplification, and saves them;

[0120] The time domain calculation module is used to perform time domain analysis and calculation on the RR interval and RR interval variation in the ECG digital signal to obtain the average heart rate of the time domain parameter The standard deviation SDNN, the root mean square RMSSD of consecutive RR intervals, and the percentage of consecutive RR intervals that differ by more than 50 ms PNN50;

[0121] Average heart rate The calculation formula is:

[0122]

[0123] Where RR i represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms; N is the length of the RR interval sequence to be analyzed;

[0124] The calculation formula for the standard deviation SDNN of the RR interval is:

[0125]

[0126] Where N is the length of the RR interval sequence to be analyzed, RR i Represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms. is the average RR interval of N heart beats;

[0127] The formula for calculating the root mean square RMSSD of consecutive RR intervals is:

[0128]

[0129] Where N is the length of the RR interval sequence to be analyzed, RR i Represents the value of the i-th RR interval in the RR interval sequence to be analyzed, in ms; RR i+1 It represents the value of the i+1th RR interval in the RR interval sequence to be analyzed, in ms, and the unit of the root mean square RMSSD is ms;

[0130] The percentage of consecutive RR intervals that differ by more than 50 ms, PNN50, is calculated as:

[0131] PNN50 = (NN50 / total number of RR intervals) x 100%

[0132] NN50 is the number of heart beats in which the difference between adjacent RR intervals is greater than 50ms among all RR intervals, in units;

[0133] The frequency domain calculation module is used to perform frequency domain analysis and calculation on the RR interval and RR interval variation in the ECG digital signal, calculate the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method, and obtain the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF of the ECG data to be analyzed by the AR model method;

[0134] The specific steps for calculating the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method are as follows:

[0135] The power spectral density PSD is calculated by performing fast Fourier transform on the RR interval data, where the power spectral density PSD can be expressed as a function of frequency, denoted as P(f), as follows:

[0136]

[0137] In the formula, x(n) (n = 0, 1, ..., N-1) represents the RR interval sequence of the heart beat, N represents the length of the RR interval sequence to be analyzed, usually N = 256 or N = 512; Δt is the average sampling interval of the RR interval sequence, that is, the average value of the RR interval sequence in this section X(f) is the discrete Fourier transform of x(n);

[0138] The power spectral density PSD is divided into the following four frequency bands: Ultra-low frequency band ULF: <0.003HZ, which is PSD ULF , Very low frequency band VLF: 0.003~0.04HZ, which is PSD VLF , low frequency band LF: 0.04~0.15HZ, which is PSD LF , high frequency band HF0.15~0.40HZ, that is PSDHF , total frequency band: ≤0.4HZ, which is PSD 总 The high-frequency band HF component in the heart rate variability signal mainly reflects the activity of the cardiac vagus nerve, and the low-frequency band LF component reflects the cardiac sympathetic nerve activity or the sympathetic vagus joint activity. Therefore, the ratio of the low-frequency band component of the signal LF / HF reflects the balance of sympathetic and vagal activities.

[0139] Heart rate variability power spectrum PSD of the total frequency band 总 =PSD ULF +PSD LF +PSD HF +PSD VLF ,

[0140] LFnorm=100×PSD LF / (PSD 总 -PSD VLF ),

[0141] HFnorm = 100 × PSD HF / (PSD 总 -PSD VLF ),

[0142] LF / HF=PSD LF / PSD HF ,

[0143] The specific steps of obtaining the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the AR model method are as follows:

[0144] The difference equation of the AR model, i.e. the autoregressive model, can be expressed as:

[0145]

[0146] Where x(n) is the RR interval sequence of the heart beat, and u(n) is a sequence with a mean of zero and a variance of σ 2 white noise sequence; p is the model order, p = 15, a k is the coefficient of the AR model, k = 1, 2, 3, …, p;

[0147] According to the autocorrelation function of x(n), the canonical equation of the AR model is constructed, namely the Yule-Walker equation. After solving it, the coefficients a are obtained. k and σ 2 The estimated value of x(n) is AR (f) can be expressed as:

[0148]

[0149] Obtaining the envelope diagram of the ECG data to be analyzed, which is a power density spectrum curve of the RR interval every 5 minutes, is the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF;

[0150] The method for selecting the determined value of the model order p is as follows: the RR interval and the RR interval variation in the ECG digital signal are analyzed and calculated in the frequency domain, the power values ​​of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function are calculated by the average period method, and the corresponding power density spectrum curve is drawn; the initial value of the model order p is set, and the envelope diagram of the power density spectrum curve of the RR interval is obtained by the AR model method, that is, the spectrum diagram of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function; if the power density spectrum curve matches the spectrum envelope diagram, the model order p is the determined value; if the power density spectrum curve does not match the spectrum envelope diagram, the value of the model order p is adjusted, and the envelope diagram of the power density spectrum curve of the RR interval is obtained again by the AR model method until the power density spectrum curve matches the spectrum envelope diagram, and the determined value of the model order p is obtained;

[0151] Abnormal judgment module, in which if the time domain parameter average heart rate If more than two of the standard deviation SDNN, the root mean square RMSSD of consecutive RR intervals, the percentage PNN50 of consecutive RR intervals with a difference of more than 50ms, and the power values ​​of the frequency domain parameters LF / HF are not within the normal threshold range, it indicates that the driver's state is abnormal; or if the spectrum envelope diagram of the ECG data to be analyzed is within the standard group diagram, it indicates that the driver's state is abnormal; if the envelope diagram of the ECG data to be analyzed is not within the standard group diagram, it indicates that the driver's state is normal.

[0152] The specific method for selecting the standard group diagram is as follows:

[0153] The ECG data from wakefulness to drowsiness are selected as standard ECG data, and the standard ECG data are input into the AR model to obtain a cluster of standard AR model graphs. Then, several graphs with a similarity greater than 0.7 to the standard AR model graphs are selected to form a standard group graph.

Claims

1. A driver safety monitoring method based on HRV autonomic nervous function parameters, characterized in that: The steps include: (1) using a sliding ring-type ECG electrode disposed on a steering wheel to collect the driver's ECG signal, the sliding ring-type ECG electrode comprising three ring-type electrode sheets disposed on the steering wheel and slidable, with an insulating isolation zone disposed between adjacent ring-type electrode sheets; (2) transmitting the collected ECG signals to an ECG amplifier for amplification, converting them into ECG digital signals after amplification and saving them; (3) Perform frequency domain analysis and calculation on the RR interval of the ECG digital signal, calculate the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method, and obtain the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF of the ECG data to be analyzed by the AR model method; The specific steps of calculating the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method in step (3) are as follows: The power spectral density PSD was calculated by fast Fourier transforming the RR interval data; The power spectral density PSD is divided into the following four frequency bands: Ultra-low frequency band ULF: <0.003HZ, which is PSD ULF , Very low frequency band VLF: [0.003,0.04)HZ, which is PSD VLF , low frequency band LF: [0.04,0.15)HZ, which is PSD LF , high frequency band HF: 0.15~0.40HZ, which is PSD HF , total frequency band: ≤0.4HZ, which is PSD 总 ; Heart rate variability power spectrum PSD of the total frequency band 总 =PSD ULF +PSD LF +PSD HF +PSD VLF , LFnorm=100×PSD LF / (PSD 总 -PSD VLF ), HFnorm=100×PSD HF / (PSD 总 -PSD VLF ), LF / HF=PSD LF / PSD HF ; The specific steps of obtaining the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the AR model method in step (3) are as follows: The difference equation of the AR model, i.e. the autoregressive model, is expressed as: ， Where x(n) is the RR interval sequence of the heart beat, and u(n) is a sequence with a mean of zero and a variance of σ 2 white noise sequence; p is the model order, a k is the coefficient of the AR model, k = 1, 2, 3, …, p; According to the autocorrelation function of x(n), the canonical equation of the AR model is constructed, namely the Yule-Walker equation. After solving it, the coefficients a are obtained. k and σ 2 The estimated value of x(n) is AR (f) is expressed as: ， Obtaining the envelope diagram of the ECG data to be analyzed, which is a power density spectrum curve of the RR interval every 5 minutes, is the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF; The method for selecting the determined value of the model order p in step (3) is as follows: performing frequency domain analysis and calculation on the RR interval of the ECG digital signal, calculating the power values ​​of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function by the average period method, and drawing the corresponding power density spectrum curve; setting the initial value of the model order p, and obtaining the envelope diagram of the power density spectrum curve of the RR interval by the AR model method, that is, the spectrum diagram of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function; if the power density spectrum curve matches the spectrum envelope diagram, then the model order p is the determined value; if the power density spectrum curve does not match the spectrum envelope diagram, then adjusting the value of the model order p, and re-obtaining the envelope diagram of the power density spectrum curve of the RR interval by the AR model method, until the power density spectrum curve matches the spectrum envelope diagram, and obtaining the determined value of the model order p; (4) If the spectrum envelope diagram of the ECG data to be analyzed is within the standard group diagram, it indicates that the driver is in an abnormal state; The method for selecting the standard group graph in step (4) is as follows: the ECG data from the process of being awake to the state of drowsiness is selected as the standard ECG data, the standard ECG data is input into the AR model to obtain a cluster of standard AR model graphs, and then the graphs with a similarity greater than 0.7 to the AR model graph of the drowsiness state are selected to form the standard group graph.

2. A driver safety monitoring system based on HRV autonomic nervous function parameters, characterized in that: It includes a sliding collar type ECG electrode, which is arranged on the steering wheel, collects the driver's ECG signal, and includes three collar type electrode pieces which are sleeved on the steering wheel and can slide, and an insulating isolation belt is arranged between adjacent collar type electrode pieces; The signal conversion and storage module transmits the collected ECG signals to the ECG amplifier for amplification, converts them into ECG digital signals after amplification, and saves them; The frequency domain calculation module is used to perform frequency domain analysis and calculation on the RR interval of the ECG digital signal, calculate the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method, and obtain the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF of the ECG data to be analyzed by the AR model method; The specific steps for calculating the power values ​​of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the average cycle method are as follows: The power spectral density PSD was calculated by fast Fourier transforming the RR interval data; The power spectral density PSD is divided into the following four frequency bands: Ultra-low frequency band ULF: <0.003HZ, which is PSD ULF , Very low frequency band VLF: [0.003,0.04)HZ, which is PSD VLF , low frequency band LF: [0.04,0.15)HZ, which is PSD LF , high frequency band HF: 0.15~0.40HZ, which is PSD HF , total frequency band: ≤0.4HZ, which is PSD 总 ; Heart rate variability power spectrum PSD of the total frequency band 总 =PSD ULF +PSD LF +PSD HF +PSD VLF , LFnorm=100×PSD LF / (PSD 总 -PSD VLF ), HFnorm=100×PSD HF / (PSD 总 -PSD VLF ), LF / HF=PSD LF / PSD HF ; The specific steps of obtaining the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF by the AR model method are as follows: The difference equation of the AR model, i.e. the autoregressive model, is expressed as: ， Where x(n) is the RR interval sequence of the heart beat, and u(n) is a sequence with a mean of zero and a variance of σ 2 white noise sequence; p is the model order, a k is the coefficient of the AR model, k = 1, 2, 3, …, p; According to the autocorrelation function of x(n), the canonical equation of the AR model is constructed, namely the Yule-Walker equation. After solving it, the coefficients a are obtained. k and σ 2 The estimated value of x(n) is AR (f) is expressed as: ， Obtaining the envelope diagram of the ECG data to be analyzed, which is a power density spectrum curve of the RR interval every 5 minutes, is the spectrum envelope diagram of the autonomic nervous function frequency domain parameters LFnorm, HFnorm, and LF / HF; The method for selecting the determined value of the model order p is as follows: the RR interval of the ECG digital signal is analyzed and calculated in the frequency domain, the power values ​​of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function are calculated by the average period method, and the corresponding power density spectrum curve is drawn; the initial value of the model order p is set, and the envelope diagram of the power density spectrum curve of the RR interval is obtained by the AR model method, that is, the spectrum diagram of the frequency domain parameters LFnorm, HFnorm, and LF / HF of the autonomic nervous function; if the power density spectrum curve diagram matches the spectrum envelope diagram, then the model order p is the determined value; if the power density spectrum curve diagram does not match the spectrum envelope diagram, then the value of the model order p is adjusted, and the envelope diagram of the power density spectrum curve of the RR interval is obtained again by the AR model method until the power density spectrum curve diagram matches the spectrum envelope diagram, and the determined value of the model order p is obtained; An abnormality judgment module, in which if the spectrum envelope diagram of the ECG data to be analyzed is within the standard group diagram, it indicates that the driver's state is abnormal; The method for selecting the standard group graph is as follows: the ECG data from wakefulness to drowsiness is selected as the standard ECG data, the standard ECG data is input into the AR model to obtain a cluster of standard AR model graphs, and then the graphs with a similarity greater than 0.7 to the AR model graph of the drowsiness state are selected to form the standard group graph.

Images