Blood pressure monitoring method based on multiple linear regression and correlation weighted selection algorithm

The invention relates to a blood pressure monitoring method based on multiple linear regression and a correlation weighted selection algorithm, and belongs to the field of digital signal processing. The blood pressure monitoring method comprises the steps of signal collection and preprocessing, feature point extraction and feature value calculation, feature type classification, mathematical model establishment, verification and output. According to the method, the detection precision is high, the change trend of the blood pressure can be accurately reflected, the correlation between a detection value and an actual value is good, and real-time and accurate monitoring of the blood pressure is achieved. Meanwhile, the blood pressure detection method is low in complexity and small in calculation amount, blood pressure can be effectively extracted from photoelectric volume pulse wave signals, the method is suitable for various existing wearable devices, high efficiency of linear regression is utilized, the extract