基于小波分析和神经网络的便携式哮喘病监测系统的校准研究

摘要

针对便携家用的哮喘病病理指标PEF、FEV1以及FEV1％监测的技术瓶颈,本文提出了一种基于小波分析和BP神经网络结合的校准方法,利用小波分析去噪提取校准样本数据,然后利用BP神经网络学习训练校准样本数据,建立校准数学模型,最后由MSP430离线实现该校准算法,该模型应用于监测系统,实现哮喘病理指标的精确监测.该校准算法克服了传统方法难以构建精确数学模型的缺点,具有收敛性好,实现方便,可广泛推广等特点.根据实验结果表明,采用该方法监测指标误差明显降低.%For solving the bottleneck of monitoring the PEF,FEV1 and FEV1％ of portable home asthma pathological indicators,this paper proposed a kind of calibration method combining with wavelet analysis and BP neural network.The wavelet analysis is used to reduce the noise and extract the calibration sample data,while the BP neural network is used to train the sample data of calibration and to establish the calibration model.The proposed calibration model was then realized by the MSP430 and was verified that it could achieve the accurate monitoring of asthma pathological indicators when it was applied to the monitoring system.The experimental results show that the error of the monitoring indicators by the proposed method is decreased obviously,demonstrating that comparing to the traditional method,the proposed algorithm can build accurate mathematical model more easily,has advantages of good convergence,easy implementation,and can be widely used in many areas.