首页 | 本学科首页   官方微博 | 高级检索  
     

表面肌电信号的时变AR模型参数估计及其在下腰痛评估中的应用研究
引用本文:曹玉珍,刘洪涛,武文君,范增飞. 表面肌电信号的时变AR模型参数估计及其在下腰痛评估中的应用研究[J]. 北京生物医学工程, 2005, 24(5): 325-328
作者姓名:曹玉珍  刘洪涛  武文君  范增飞
作者单位:天津大学精密仪器与电子工程学院,天津,300072;天津大学精密仪器与电子工程学院,天津,300072;天津大学精密仪器与电子工程学院,天津,300072;天津大学精密仪器与电子工程学院,天津,300072
摘    要:本文针对表面肌电信号的非平稳特性,采用时变参数AR模型对表面肌电信号进行分析,将线性非平稳问题转化为线性时不变问题,并采用递推最小二乘算法求解模型的时变参数.在此基础上,提出了结合奇异值分解进行参数优化,进而进行模式分类的方法.能够成功地区分下腰痛患者治疗前后的状态,为下腰痛的诊断、治疗与康复判定奠定了一定的基础.

关 键 词:表面肌电信号  时变参数AR模型  奇异值分解  模式分类
文章编号:1002-3208(2005)05-0325-04
收稿时间:2004-04-28
修稿时间:2004-04-28

The Time-Varying Model Parameter Estimation and Its Application to Low Back Pain Evaluation
CAO Yuzhen,LIU Hongtao,WU Wenjun,FAN Zengfei. The Time-Varying Model Parameter Estimation and Its Application to Low Back Pain Evaluation[J]. Beijing Biomedical Engineering, 2005, 24(5): 325-328
Authors:CAO Yuzhen  LIU Hongtao  WU Wenjun  FAN Zengfei
Affiliation:College of precision instrument and opto-electronics engineering, Tianjin University, Tianfin 300072
Abstract:A surface EMG signal identification method based on time-varying autoregressive model is presented. To fully utilize the nonstationary character of the EMG signal, time-varying AR model is employed to get the signal's frequency representation. Singular value decomposition is then used to extract feature vector for pattern identification. Experiments from patients shows that the method is stable and efficient for extracting features. It would promote practical applications in the field of rehabilitation medicine.
Keywords:surface EMG time-varying autoregressive model SVD pattern recognition
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号