摘要
为了提高胎儿心电(FECG)提取的准确性和效率,研究了采用最小均方算法、递归最小二乘(RLS)算法、仿射投影算法、能量归一化最小均方算法的四种自适应滤波器应用于FECG提取的速度和准确性。根据设计的基于MATLAB的FECG提取算法仿真分析了四种自适应滤波器的收敛速度与稳态性能的影响因素,并通过调整和优化FECG提取参数对四种自适应滤波器进行了仿真对比。研究结果表明,RLS算法自适应滤波器性能最优,可以快速准确地从母体心电中分离出FECG。
In order to improve the accuracy and efficiency of fetal electrocardiogram（FECG）extraction,the speed and accuracy of four kinds of adaptive filters using least mean square algorithm,recursive least squares（RLS）algorithm,affine projection algorithm and energy normalized least mean square algorithm respectively for FECG extraction are studied.The influencing factors of the convergence rates and steady-state performances of the four kinds of adaptive filters are simulated and analyzed according to the FECG exaction algorithm designed on MATLAB.A comparative study on the four kinds of adaptive filters is done by adjusting and optimizing the parameters of the FECG extraction.The results indicate that the RLS adaptive filter has the best performance,and can extract FECG from maternal electrocardiogram fast and accurately.
出处
《南京理工大学学报》
EI
CAS
CSCD
北大核心
2011年第6期847-852,共6页
Journal of Nanjing University of Science and Technology
基金
中国博士后科学基金面上项目(2011M500896)
南京邮电大学青蓝基金(NY210047)
东南大学生物电子学国家重点实验室开放研究基金
东南大学MEMS教育部重点实验室开放研究基金
关键词
胎儿心电
自适应滤波器
最小均方算法
递归最小二乘算法
仿射投影算法
能量归一化最小均方算法
母体心电
fetal electrocardiogram
adaptive filter
least mean square algorithm
recursive least squares algorithm
affine projection algorithm
energy normalized least mean square algorithm
maternal electrocardiogram
