健康监护系统中基于最长公共子序列的用户行为分析
目前已经存在监测用户行为的系统，但存在侵犯用户隐私的问题。因此，在本文中，我们并不采用视频或音频设备，而是使用简单、便宜的红外传感器收集行为事件，并且仅仅判断行为是否异常，而不需要判断用户的每个行为究竟是在做什么。9526
在实验中，我们通过最长公共子序列比较的方法可以判断出收集到的用户行为序列是否异常，及时报知监护人员做出恰当的处理。同时，也研究了3种不同最长公共子序列算法的复杂度，实验证明使用基于矩阵搜索的动态规划算法是最佳的。本课题提出的采用简单的相似度比较检查异常行为的方法可以广泛的适用于家庭、医疗场所的监护系统中。
关键词 无线传感网络 健康监护系统 字符串匹配 最长公共子序列 异常行为检测
毕业设计说明书（论文）外文摘要
Title Analysis of User Behavior Based on The Longest Common Subsequence in Health Care System Abstract
As the increasing number of the elderly, more and more elderly people are likely to have a sudden behavioral change due to their aging or existing health problems,so there must need someone take care of them all day long. However, because most modern people have the problem of working pressure or little leisure time, they cannot take care of the elderly all-day, so when an accident occurs, it is likely to miss the best time of rescuing. Therefore, it is necessary to have an autonomous system that can monitor them in order to prevent emergent situation in advance. In this paper, we study human behavioral patterns of elderly who lives alone in a health monitoring system based on wireless sensor networks. We try to collect series of events for a person's behavior every day. And then, use the longest common subsequence algorithm to detect the abnormal behavior. Once an exception occurs,the alarm should be immediately reported to family members or guardians.
At present, there is already existing system to monitor user's behavior, but it has the problem of invasion of user's privacy. Therefore, we do not use video or audio equipment, but, use the simple, inexpensive infrared sensors to collect behavioral event. And only to determine whether it is abnormal behavior or not, without the need to determine what the user's behavior exactly is.
In the experiment, we determine whether the user's behavior sequence that collected by wireless sensor networks is normal by using the algorithm of the longest common subsequence, then notify the custody officer to make the appropriate treatment. At the same time, we also study the complexity of three algorithms to identify the longest common subsequence, and the experiment shows that it is the best to use the dynamic programming algorithm which is based on the matrix search. The topic proposed in this paper using a simple similarity measurements to check for abnormal behavior can be widely applied in the monitoring system of the family and medical establishments.
Keywords wireless sensor network system Heath care system String matching the longest common subsequence monitoring of the abnormal behavior