体温实时监测系统设计  

Design of human temperature real-time monitoring system

作　　者：胡良文 陈向东[1] 吴坤 马景 HU Liang-wen;CHEN Xiang-dong;WU Kun;MA Jin(School of Information Science&Technology,Southwest Jiaotong University,Chengdu 610031,China)

机构地区：[1]西南交通大学信息科学与技术学院,四川成都610031

出　　处：《电子设计工程》2019年第1期140-143,148,共5页Electronic Design Engineering

基　　金：四川省科技支撑计划项目(2016JZ0028)

摘　　要：针对医院对住院患者人体基础体温高精度、连续24小时实时监测的需求。采用ZigBee无线通信技术,实现多节点的自动组网与数据传输;利用4X1DS18B20温度传感器阵列采集温度数据并对该组数据进行算术平均运算,结合最小二乘法数据拟合算法,提高体温测量精度;通过利用支持向量机分类算法结合卡尔曼滤波算法,实现对干扰脉冲的有效滤除。通过与传统水银温度计体温测量、手持式电子体温计体温测量对比试验,验证了该系统连续测量的可靠性,在人体主要体温测量段35～39℃,测量精度为0.06℃。Aimed at the hospital's need for high-precision,continuous24-hour real-time monitoring of the patient's basal body temperature.ZigBee wireless communication technology is used to realize multinode automatic networking and data transmission;4X1DS18B20temperature sensor array is used to collect temperature data and the group of data is arithmetically averaged,combined with least-squares data fitting algorithm to improve the accuracy of body temperature measurement;Using the support vector machine classification algorithm combined with the Kalman filter algorithm to achieve effective filtering of interference pulses.Compared with the traditional mercury thermometer body temperature measurement and the hand-held electronic thermometer body temperature measurement comparison test,the reliability of the continuous measurement of the system was verified.The main body temperature measurement section of the human body was35~39°C,and the measurement accuracy was0.06°C.

关 键 词：人体体温检测 ZIGBEE 卡尔曼滤波 支持向量机 DS18B20阵列 

分 类 号：TN98[电子电信—信息与通信工程]