基于WiFi的多目标呼吸检测方法研究  

Research on Multi-Target Respiratory Detection Method Based on WiFi

作　　者：秦娥[1] 张锦鸿 徐也淳 贾逸哲 王可 毛科技[1] 李卫锋 邱杰凡 何文秀 QIN E;Zhang Jinhong;XU Yechun;Jia Yizhe;Wang Ke;MAO Keji;LI Weifeng;QIU Jiefan;HE Wenxiu(College of Computer Science and Technology,Zhejiang University of Technology,Hangzhou Zhejiang 310023,China;School of Public Administration,Zhejiang University of Technology,Hangzhou Zhejiang 310023,China;Watone Cloud Data Co.,Ltd,Hangzhou Zhejiang 310030,China;Zhijiang College,Zhejiang University of Technology,Shaoxing Zhejiang 312030,China)

机构地区：[1]浙江工业大学计算机科学与技术学院,浙江杭州310023 [2]浙江工业大学公共管理学院,浙江杭州310023 [3]浙江华通云数据科技有限公司,浙江杭州310030 [4]浙江工业大学之江学院,浙江绍兴312030

出　　处：《传感技术学报》2025年第1期112-121,共10页Chinese Journal of Sensors and Actuators

基　　金：浙江省基础公益研究计划项目(LGG22F020014);国家自然科学基金项目(62072410)。

摘　　要：在探讨多目标呼吸监测技术的过程中,挑战主要集中于如何准确无误地分离和识别多个对象的呼吸信号。解决这一问题的关键技术是在WiFi系统中采用MIMO技术,该技术通过增加物理测量点的数量,使得在静态环境中对多个人的呼吸进行监测成为可能。目前,多人呼吸监测的方法大多依赖于对信道状态信息(CSI)的幅度进行频谱分析,以便提取出呼吸信号。然而,依赖于频谱分析的方法面临两个主要问题:首先,所能达到的频率分辨率受限于数据的持续时间,这意味着在数据长度不足时,无法区分频率相近的呼吸信号;其次,频谱分析只能提供一定时间段内的平均呼吸频率,不能详细反映出这段时间内呼吸模式的变化。为了解决这些问题,开发了一种名为MoRadio的新型多人呼吸监测方法。通过实验验证,证明该方法能够有效识别和监测呼吸率相近的多个目标,且在多人环境下对目标数量的识别准确率高达95%,bpm最佳可达0.1。The challenge in multi-target respiratory detection lies in effectively and accurately extracting respiratory signals from multiple targets.To address this,the use of MIMO technology in WiFi systems is essential.By increasing the number of physical observation points,the detection of breathing in multiple individuals in a static environment is achieved.Existing methods for multi-person respirato-ry detection primarily utilize spectral analysis of channel state data amplitude to obtain target respiratory information.However,respira-tory detection based on spectral analysis faces two main issues:①The frequency resolution achievable by spectral analysis depends on the length of the data.Without sufficiently long data,it is challenging to distinguish closely spaced respiratory signals.②Spectral analy-sis can only provide an average respiratory rate over a period,failing to capture specific changes in breathing patterns during that time.To overcome these problems,a multi-person respiratory detection method,MoRadio,is introduced,which is capable of detecting targets with similar respiratory rates.Experimental results show that the proposed method achieves an accuracy of 95%,and 0.1 bpm in identif-ying the number of people in multi-person environments.

关 键 词：多目标呼吸检测 MIMO技术 频谱分析 信道状态数据 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]