摘要
针对老年人健康监护与安全保障的迫切需求,本研究提出一种基于STM32微控制器的智能手环系统,集成一键紧急呼叫与按需回访功能,构建多模态健康监测与实时安全防护体系。系统通过DS18B20温度传感器、MAX30102心率/血氧传感器实现生理参数的精准采集,结合GPS/北斗双模定位模块与4G通信技术,实现地理位置信息与健康数据的云端同步。OLED显示屏提供本地化数据可视化,家属可通过定制化移动端APP远程获取老人实时体温、心率、血氧饱和度及位置轨迹,形成“监测–预警–响应”闭环管理机制。在紧急场景下,用户触发物理按键后,系统自动解析位置坐标并通过蜂窝网络向预设联系人发送SOS警报信息。经实验室测试与实地验证,温度监测误差 ≤ ±0.3℃ (−10℃~85℃),心率检测准确率达98.6%,定位精度优于5 m,4G模块丢包率低于0.8%。研究表明,该设计突破了传统养老设备功能单一的技术瓶颈,通过异构传感器融合与低功耗优化策略(待机电流 < 10 μA),为智慧养老提供了可扩展的技术框架,未来可通过接入AI健康分析模型进一步强化疾病预测能力。
To address the urgent needs of health monitoring and safety assurance for the elderly, this study proposes an intelligent bracelet system based on an STM32 microcontroller, integrating one-touch emergency calling and on-demand callback functions to establish a multimodal health monitoring and real-time safety protection framework. The system employs a DS18B20 temperature sensor and MAX30102 heart rate/blood oxygen sensor to achieve precise physiological parameter acquisition. Combined with a GPS/BeiDou dual-mode positioning module and 4G communication technology, it enables cloud synchronization of geographical location information and health data. An OLED display provides localized data visualization, while a customized mobile app allows family members to remotely access real-time body temperature, heart rate, blood oxygen saturation, and location trajectories of the elderly, forming a closed-loop “monitoring-alert-response” management mechanism. In emergency scenarios, triggering the physical button automatically parses location coordinates and sends SOS alerts via cellular networks to preset contacts. Laboratory tests and field verification demonstrate a temperature monitoring error of ≤±0.3˚C (−10˚C~85˚C), heart rate detection accuracy of 98.6%, positioning precision better than 5 meters, and 4G module packet loss rate below 0.8%. The research indicates that this design overcomes the technical limitations of single-function traditional elderly care devices. Through heterogeneous sensor fusion and low-power optimization strategies (standby current < 10 μA), it provides an extensible technical framework for smart elderly care. Future work could enhance disease prediction capabilities by integrating AI-based health analysis models.
出处
《传感器技术与应用》
2025年第4期632-639,共8页
Journal of Sensor Technology and Application
基金
中央民族大学大学生创新训练计划项目(URTP2024110876)。
