一类生理信号的采集、建模和辨识

Sampling, Model Building and Identification of a Category of Physiological Signals

中文关键词  生理信号  时间序列分析  建模  参数辨识

英文关键词  physiological signal  time series analysis  model building  parameter identification

中文摘要

      本文对生理信号的采集、建模和辨识作了较系统的探讨。在生理信号的采集方面，介绍了实时和离线信号的采集的设计思想和实现方案。在生理信号的数据建模方面，引入了ARMA 建模方案，着重研讨了模型的实现途径、参数辨识和平稳化方法，编制建模程序及实现框图。在心功能病类辨识方面，对典型心功能信号——ECG、CAG和ACG，根据类比原则，利用模型分析结果，对三类心血管病——高血压、冠心病和心肌炎作了聚类辨识，取得用参数域值辨识病理的结果。

英文摘要

      This paper intends to give a systematic investigation to the sampling, model building and identification of a category of physiological signals. As to model fitting of physiological signals, by applying the ARMA model scheme to the sampled cardiac function data series, we stressed on the realistic methodology, the parameter identification technique and stationarized method. By virture of typical cardiac function signals-ECG, ACG,CAG, combining with microscopic analysis results, we have made characteristic identification microscopically to the typical desease-hypertension, coronary heart desease and myocarditis, according to the parameter `s domain values of ARMA models.