摘要
在床垫式睡眠监测系统中,呼吸波的真实形态容易在显著体动发生时被掩盖。本文提出一种显著体动下的呼吸波补偿算法。首先利用窗口阈值判别法对床垫系统得到的上身原始信号进行分析,得到体动标识;之后设计模板补偿算法对显著体动下的呼吸波进行补偿。补偿后的信号能够有效反映受试者睡眠期间的呼吸状态,大大减小了呼吸事件的误判率,有效率可达91.36%。多次实验表明,该算法简单可行,实现了呼吸波的有效获取和睡眠呼吸事件的纠正。
In MSMSMS, the actual shape of respiration signal is easily covered up during notable body movements. A compensation algorithm is presented for the respiration signal during notable body movements. Firstly, the upper signals from MSMSMS are analyzed with window threshold method, and the body movements are classified and rec- ognized. Then, the respiration signal during notable movements are compensated with template compensation method. The compensated signal can reflect the actual respiration state efficiently, therefore the error rate of breathing events is decreased a lot. The efficiency rate can reach to 91.36%. The experiment results show that this algorithm is easy to op- erate, and can realize the efficient extraction of respiration signal and the correction of sleep-related breathing events.
出处
《电子测量与仪器学报》
CSCD
2009年第4期55-59,共5页
Journal of Electronic Measurement and Instrumentation
