摘要普适计算是信息空间与物理空间的融合。基于普适计算的健康监护系统是普适计算思想在健康和医疗领域的一种重要应用,目前已经得到了广泛的关注。在数据融合方面,由于单一类型传感器只能提供有限的数据,而不同种类的传感器可以相互补充,并收集到更多的关于健康状况方面的信息。所以,多传感器融合在无线健康监控方面十分重要。可以将多个传感器采集到的不同种类的数据,看作是多个随机变量。在处理多随机变量的推理融合问题中,贝叶斯网络理论可以很好地解决此类问题。在健康监护和医疗诊断中使用贝叶斯网络进行推理预测,同时,使用无线传感网络技术,完成对受监护人进行远程监护。本文在介绍传感器融合理论的基础上,通过老人跌倒异常动作检测试验,完成贝叶斯网络的诊断推理过程。61638
毕业论文关键词 普适计算 多传感器 医疗诊断 数据融合 贝叶斯网络
Title The Study of Sensor Data Fusion Problem in Pervasive Health Care Monitoring System
Abstract Pervasive computing is the integration of information space and physical space, the health care system based on pervasive computing, is the ideas of the pervasive computing. This system is an important application in the field of health and medical care. Now the system has been widely concerned. In terms of data fusion, as a result of a single type of sensor can only provide limited data, and the different kinds of sensors can complement each other, and generate more comprehensive health information. Therefore, multiple sensor fusion is very important in wireless health monitoring. Different sensors collect different data which can be looked as multiple random variables. In dealing with multiple random variables reasoning fusion, the theory of Bayesian network is a good way to solve such problems. In health care and medical diagnostics using Bayesian network inference prediction, using wireless sensor network, to accomplish remote monitoring. Based on the sensor fusion, it complete the diagnosis reasoning process of Bayesian network with the old man fell and abnormal motion detection experiment.
Keywords pervasive computing multiple sensor medical diagnosis data fusion Bayesian network
目 次
1 绪论 1
1.1 背景介绍 1
1.3 本文主要研究内容 4
2 相关技术研究发展现状 5
2.1 数据融合技术 5
2.2 贝叶斯网络 5
2.3 贝叶斯网络的优点 6
2.4 静态BN和动态BN 7
2.5 贝叶斯网络的性质及推理过程 10
2.6 本章小结 12
3 贝叶斯网络理论在普适医疗系统中的应用 12
3.1 无线传感器网络在医疗诊断中的应用 12
3.2 对老人异常行为的检测分析 15
3.3 使用贝叶斯网络对老人异常行为检测 16
3.4 系统总体概要 16
3.5 本章小结