关键词: 无损酒精检测, 近红外光谱法, 指尖信号, 卡尔曼滤波器, 三样条插值法

Abstract: In order to solve the problem of droplet transmission risk of breath alcohol detector, a portable and pollution-free blood alcohol concentration detection scheme based on near-infrared spectroscopy is proposed. we successively completed the signal acquisition, amplification, filtering and re-amplification with hardware, and obtained the standard pulse wave signal with software algorithms such as Kalman filtering and three-sample interpolation method to remove the base. Later, the mathematical model is established in the upper computer by the measurement results of the expiratory type and the amplitude of the pulse wave signal, and the mathematical model is written into the microcontroller to realize the offline measurement. The accuracy of the mathematical model is tested through validation experiments. The results show that it meets the accuracy requirement, the relative error is less than 10% and meets the practical requirements.

Key words: non-destructive alcohol detection, near infrared spectroscopy, fingertip signal, kalman filter, three- sample interpolation