基于变步长凸组合LMS自适应滤波高原动态血氧检测系统设计

Design of Motion SpO_(2) Detecting System based on VSCLSM Algorithm

针对基于光电检测方法的血氧仪只能在静止状态下测试血氧饱和度(血氧)的问题,提出一种基于变步长凸组合LMS自适应滤波动态血氧检测方法。方法对原始光电描记脉搏波信号经预处理后,通过固步长与变步长LMS自适应算法组合,以三轴加速度计采集信号作为运动噪声参考信号,获得理想脉搏波信号再准确计算血氧值。在此基础上,开发了一套基于物联网的穿戴式高原血氧检测系统,经过模拟正常行走及在不同海拔条件下测试,最大误差不超过4%,验证了算法的有效性和准确性。结果显示,基于物联网的血氧监测系统对于高原旅游中高原反应的实时监测与提醒具有实际意义。

Aiming at the issue that SpO_(2) measurement by photoelectric pulse oximetry can only be measured accurately in a static state,an algorithm based on variable step-size convex combination LMS adaptive filtering(VSCLMS)for SpO_(2) in a dynamic state is proposed.Firstly,a photoplethysmography(PPG)signal is acquired in a motion state and preprocessed by low-pass filtering.Then,using the tri-axial accelerometer as the reference signal of motion noise,the ideal PPG signal is obtained through the VSCLMS adaptive algorithm and SpO_(2) values are acquired.Finally,a motion SpO_(2) detecting system for altitude reaction based on the wearable devices and the Internet of Things is developed.After simulating normal walking state at a different altitude,the error is less than 4%,which verify the effectiveness and accuracy of the proposed algorithm.The motion SpO_(2) detecting system is of practical significance for real-time monitoring and reminding of the altitude reaction in plateau tourism.