基于FMCW雷达的呼吸模式分类技术研究

Research on Respiratory Pattern Classification Technology Based on FMCW Radar

针对当前使用调频连续波雷达的呼吸模式分类算法准确度不高的问题,本文提出一种基于一维卷积神经网络(1DCNN)结合长短时记忆(LSTM)网络的多呼吸模式分类方法。方法共分为四步:第一步,对雷达提取的呼吸信号进行预处理;第二步,使用快速傅里叶变换(FFT)与连续小波变换(CWT)提取呼吸信号特征;第三步,根据呼吸特征对五种呼吸模式信号(正常呼吸、呼吸过速、呼吸过缓、呼吸深大、呼吸暂停)打标签制作数据集;第四步,使用数据集训练网络得到模型,并使用新数据测试模型。实验结果表明,此方法分类准确度要比现有使用CNN网络方法高5%左右。

Aiming at the low accuracy of respiratory pattern classification algorithms used in frequency modulated continuous wave(FMCW)radar,this paper proposes a multi respiratory pattern classification method based on one⁃di⁃mensional convolutional neural network(1DCNN)combined with long and short term memory network(LSTM).The method has four steps:the first step is to preprocess the respiratory signal extracted by radar;the second step is to ex⁃tract the characteristics of respiratory signals by using fast fourier transform(FFT)and continuous wavelet transform(CWT);the third step is to label the five respiratory pattern signals(normal breathing,tachypnea,bradypnea,deep breathing and apnea)according to the respiratory characteristics to form a data set;the fourth step is to use the data set to train the network to get the model,and use the new data to test the model.The experimental results show that the clas⁃sification accuracy of this method is about 5%higher than that of the existing CNN network methods.