摘要
室内人体朝向检测是室内感知的一个重要分支,在监控安防、体感游戏、智能家居等领域都有广泛的应用。提出了一种新的人体朝向检测方案Wi O:采用无线局域网中的信道信息作为人体朝向检测的传感数据,在此基础上通过机器学习算法进行分类,取得了良好的朝向检测效果。在朝向检测可靠性方面,同时考虑了该传感信号的幅度和相位信息的基础上综合了多对天线对的结果,并研究了不同环境因素对检测结果的影响。实验表明所提出的方法能有效检测人体朝向,最佳检测准确度可以达到98%。
Human orientation detection is an important branch of human sensing,which have a wide application in many fields such as security,smart housing and somatic games. A new human body orientation detection algorithm WiO was presented： utilize wireless channel information in Wireless Local Area Network to act as the physical quantity; we adopt machine learning algorithm to classify the data into the according location,which obtains a relatively good performance of classification; In order to further improve the reliability of detection,we combined the wireless channel amplitude and phase information,and integrate multiple antenna pairs; Meanwhile,we consider the impact of different environmental factors on the experimental results. The experiment shows that our method detects the human orientation effectively,the classification accuracy of four directions in best environment is 98%.
作者
吴哲夫
李嘉南
王中友
毛科技
宣琦
翔云
WU Zhefu;LI Jianan;WANG Zhongyou;MAO Keji;XUAN Qi;XIANG Yun(Collage of Information Engineering, Zhejiang University of Technology, Hangzhou 310023, China;Zhejiang Communications Service Zhejiang Company, Hangzhou 310050, China;Collage of Computer, Zhejiang University of Technology, Hangzhou 310023, China)
出处
《传感技术学报》
CAS
CSCD
北大核心
2018年第1期90-96,共7页
Chinese Journal of Sensors and Actuators
基金
浙江省自然科学基金项目(LY18F010025
LY18F030021)
国家自然科学基金项目(61572439)
