Analysis Of The Characteristic Parameters Of Human Pulse Wave Under The Condition Of Slight Drunkenness

Drunk driving is an important factor to affect traffic safety.At present,the commonly used drunk driving detection methods such as exhalation method and blood drawing method can’t meet the requirements of fast and accurate judgment at the same time.Study has shown that the physiological changes of human body after drinking can be expressed by pulse wave.According to this thesis,a new method of drunk driving detection based on the analysis of characteristic parameters of human pulse wave is proposed,which uses the photoelectric volume pulse wave sensor to collect the pulse wave of the human body and explore the variation of human pulse wave characteristic parameters under the condition of slight drunkenness,the main research contents are as follows:(1)The status quo of drunk driving detection technology at home and abroad is reviewed,and the characteristics of pulse wave signal are analyzed.The thesis demonstrated the feasibility of using pulse wave signal to detect drunk driving.The method of dividing and identifying pulse wave feature points is expounded,and the time domain analysis method and time-frequency analysis method are emphasized.(2)The thesis designed the pulse wave signal detection system,and introduced the design of the hardware circuit,the lower computer software and the Lab VIEW interactive interface of the host computer.The system uses the infrared transmission pulse wave sensor HKG-07B to collect the pulse wave of the human fingertip.The LM324 chip is used to replace the traditional complex circuit to adjust the pulse wave signal output by the sensor,and then input to the upper computer for subsequent processing.(3)The processing method of pulse wave signal is studied.The sym8 wavelet base is used to denoise the pulse wave signal,the extreme value method is used to identify the period,the cubic spline interpolation method is used to remove the baseline,and then the six characteristic points of the pulse wave are sequentially identified.For the inconspicuous tidal wave,the position of the feature point e is determined by analyzing the difference between the 4th and 5th order coefficients after wavelet decomposition.The recognition accuracy of the obvious e point is 97.5%,and the recognition accuracy of the unobvious e point can be Up to 84%.(4)The human body drinking test was conducted using the designed pulse wave signal detection system.Thirty healthy college students were selected as experimental subjects.At the same time,the concentration of breath alcohol was detected by a breath-type alcohol detector.Finally,the changes of characteristic parameters in the state of non-drinking and drinking were compared.The analysis found that after drinking alcohol,the left ventricular ejection function increased,the vascular elasticity coefficient decreased,the peripheral resistance became lower,the left ventricular systolic and diastolic phases became shorter,the pulse wave tidal wave gradually disappeared,and the amplitude of the re-pulse wave became larger.The characteristic parameters Q2,Q3,and Q4 all change correspondingly with the change of human alcohol concentration.This thesis explores the changes of the pulse wave of the human body after drinking,and lays a theoretical and practical foundation for the next step to detect drunk driving by pulse wave.