基于 SVM 的便携式睡眠监测系统设计

摘要

Objective Sleep monitoring is an important part of the analysis of sleep quality , yet the sleep monitoring system available now is complex and cumbersome .A portable sleep monitoring system based on support vector machines ( SVM) is proposed in this paper with great convenience and efficiency .Methods The system’ s hardware consists of the server and the user equipment .The user equipment with high portability is used for data acquisition and data transmission . The server is used for data analysis and resource maintenance.SVM is adopted as the automatic sleep analysis algorithm in the server .Based on extracted features, sleep stages are got with directed acyclic graph as the multi-classification method.Results The research results based on patient EEG analysis show that the system can reach a high accuracy rate and take short analysis time average analysis time of 1.45 seconds.Conclusions The compact user equipment is highly portable , and it can feedback the correct result to the users in real time , thus confirming that the design has a promising future in sleep monitoring .%目的：睡眠监测是睡眠质量分析中重要的环节，但目前的睡眠监测系统复杂而且难以携带。本文提出基于支持向量机的便携式睡眠监测系统，以方便地实时监控睡眠。方法该系统硬件部分由服务器和用户端设备构成，其中用户端设备负责数据采集和数据传输，服务器端负责数据分析及相关的资源管理。睡眠分析软件采用支持向量机（ support vector machines ， SVM）作为分析算法，在提取特征值的基础上，以有向无环图作为多分类策略分析得到睡眠的时相。结果对于患者的睡眠脑电实验表明分析正确率高，所需的分析时间短。结论该系统用户端设备体积小，方便携带，分析正确率高，实时性好，在睡眠监测领域具有良好的应用前景。