基于自适应RLS算法的非接触式生命参数检测中呼吸和心跳信号的分离

Separation of Respiration and Heartbeat Signal in Non-contact Life Parameters Detecting Based on RLS Algorithm

目的:从生物雷达非接触式检测所得生命参数信号中提取心跳信号,实现呼吸和心跳信号的有效分离。方法:首先用生物雷达采集系统采集由呼吸引起的携带有心跳等信号的体表微动信号,其次对采集的信号进行预处理,滤除高频噪声,并放大信号,然后运用自适应递归最小二乘(RLS)算法进行处理。结果:自适应RLS算法能够有效地抑制生命参数信号中的呼吸成分,提取出心跳信号。结论:自适应RLS算法可以自动调整自身参数以逐步实现最优处理结果,这对于处理随机性强的生物医学信号十分有效;此外,算法收敛速度快,便于信号实时处理,对临床监护十分有用。

Objective To extract heartbeat signal from life parameters signal which is detected by bio-radar with non-contact way, in order to separate respiration signal and heartbeat signal efficiently. Methods Firstly, heartbeat signals and others surface micro-signals were colleted by bio-radar. Secondly, the signals were pretreated by eliminating high frequency noise and magnifying the signal, and then the way of adaptive RLS algorithm was used in the process. Results The adaptive RLS algorithm could restrain respiration component of vital parameter signal and extract heartbeat signal. Conclusion The adaptive RLS algorithm could regulate the parameters of itself automatically in order to achieve optimal outcome, and it＇s useful to process particular stocilastic biomedical signal. The convergence rate of algorithm is very fast and is convenient to make signal real-time processing and particularly useful for clinical-monitor. [Chinese Medical Equipment Journal, 2009,30（3）：25-27]