基于神经网络的WiFi睡眠呼吸暂停智能监测技术

为了克服传统睡眠呼吸监测方案未考虑实际受测人员在测试区域可能存在呼吸暂停、正常呼吸或者离开测试区域的问题，设计一种基于家庭WiFi的睡眠呼吸暂停智能监测系统。利用线性拟合消除接收天线的信道状态信息 (channel state information， CSI) 相位误差，并利用小波变换去除信号幅值的噪声；结合短时傅里叶变换和滑动窗口法对信号进行分割；提取天线间相位差的方差等特征并利用神经网络模型对呼吸暂停进行识别，排除睡姿变化带来的干扰。实验结果表明，该系统对于呼吸暂停的检测率达到95.6%以上，能够作为日常的呼吸暂停监测方案并为用户提供健康参考。

WiFi sleep apnea intelligent monitoring technology based on neural network

In order to overcome the problem that the traditional approaches of breathing monitoring hardly consider that actual subject may be apnea, breathing normally or leaving in the test area, an intelligent sleep apnea monitoring system based on home WiFi is proposed in this paper. Firstly, linear fitting and wavelet transform are used to eliminate the phase errors of channel state information (CSI) of the receiving antennas and remove the noise of signal amplitude, respectively. Then, we segment the signal by combining the short-time Fourier transform and sliding window method. Finally, we extract the features such as the variance of the phase difference between antennas, and build the neural network model to identify apnea state to eliminate interference caused by changes in sleep postures. The experimental results show that the detection rate for apnea is over 95.6%. The proposed system can be a daily apnea monitoring approach and provide health reference for users.