Research on Activity-based Device-Free Human Identification Approach
首发时间:2021-04-08
Abstract:In recent years, with the development of device-free sensing (DFS), researchers have applied it to the field of identity recognition. By taking advantage of the unique influence that different people have on surrounding wireless signals, the technology is able to identify people in a contactless way. Some progress has been made in identification by analyzing the gait information in the received wireless signals, but it needs to provide enough space for walking, which limits its application scenarios to some extent. In order to solve this problem, this paper proposes WAID, a recognition system based on human activities. Through analyzing the unique influence of different people doing the same activity on the channel state information (CSI) of WiFi signal, the identity information of people can be extracted from it. Thus, identification can be achieved by performing activities in fixed locations where there are no area requirements. The experimental results show that in the case of 2 to 6 people, the average recognition accuracy of WAID is 94.3% to 88%, and the average accuracy under six positions is 84.8%.
keywords: DFS Human Identification WiFi signals CSI
点击查看论文中文信息
基于人体活动的无设备身份识别方法研究
摘要:近年来,随着无设备感知技术的发展,其在身份识别领域中受到广泛关注。通过利用不同人对周围无线信号的独特影响,该技术能够以非接触的方式识别身份。通过分析接收到的无线信号中的步态信息,基于步态的身份识别取得了一定的进展,但它需要为步行提供足够的空间,这在一定程度上限制了它的应用场景。为了解决这一问题,本文提出了WAID,一种基于人体活动的识别系统,通过分析不同人做同一活动对WiFi信号中信道状态信息造成的独特影响,从中提取出人的身份信息。因此,可以通过在没有区域要求的固定位置进行活动来实现身份识别。实验结果表明,2~6人的条件下,身份识别准确率为94.3%~88%,6个位置的条件下平均准确率为84.8%。
引用
No.****
动态公开评议
共计0人参与
勘误表
基于人体活动的无设备身份识别方法研究
评论
全部评论0/1000