||
本文为加拿大卡尔顿大学(作者:Zachary James Baird)的硕士论文,共113页。
由于雷达具有安全、非接触、保护患者隐私等优点,在长期护理中被提出用于监测老年患者的健康状况。随机身体运动(RBM)使得雷达回波信号费解,精确估计生命体征变得困难,甚至不可能。本文将人体活动分类作为RBM的一个预处理步骤。提出了姿态分类的概念,以协助预防跌倒。本文研究了两种常用的雷达结构:连续波多普勒和超宽带。连续波和超宽带的人体活动分类平均准确率分别为92%和86%。在连续波和超宽带条件下,姿态分类的平均精度分别为64%和85%。提出了一种适用于UWB的占用率检测算法,平均精度达到88%。本文的贡献在于提出了一种对两种雷达类型都能处理运动目标的分层处理方法。
Radar has been proposed for monitoring the health of elderlypatients in long term care because it is safe, non-contact and preserves theprivacy of patients. Random body movements (RBM) obscure radar return signalsmaking it difficult if not impossible to accurately estimate vitals. Activityclassification is presented in this thesis as a preprocessing step for dealingwith RBMs. Posture classification is presented in this thesis for assistance inpreventing falls. Two popular radar architectures- continuous wave (CW) Dopplerand ultra-wideband (UWB) are investigated in this thesis. Activity classificationis performed with 92% average accuracy with CW and 86% with UWB. PostureClassification is performed with 64% average accuracy with CW and 85% with UWB.An occupancy detection algorithm was also developed for UWB and achieved 88%average accuracy. The contribution of this thesis is a proposed hierarchical processingapproach for both radar types capable of dealing with moving subjects.
更多精彩文章请关注公众号:
Archiver|手机版|科学网 ( 京ICP备07017567号-12 )
GMT+8, 2024-10-20 03:44
Powered by ScienceNet.cn
Copyright © 2007- 中国科学报社