Multidomain Correlation-Based Multidimensional CSI Tensor Generation for Device-FreeWi-Fi Sensing  

作　　者：Liufeng Du Shaoru Shang Linghua Zhang Chong Li JianingYang Xiyan Tian 

机构地区：[1]School of Mechanical and Electrical Engineering,Henan Institute of Science and Technology,Xinxiang,453003,China [2]School of Communications and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing,210003,China [3]School of Mechatronic Engineering and Automation,Shanghai University,Shanghai,200444,China

出　　处：《Computer Modeling in Engineering & Sciences》2024年第2期1749-1767,共19页工程与科学中的计算机建模（英文）

基　　金：the National Natural Science Foundation of China under Grant 61771258 and Grant U1804142;the Key Science and Technology Project of Henan Province under Grants 202102210280,212102210159,222102210192,232102210051;the Key Scientific Research Projects of Colleges and Universities in Henan Province under Grant 20B460008.

摘　　要：Due to the fine-grained communication scenarios characterization and stability,Wi-Fi channel state information(CSI)has been increasingly applied to indoor sensing tasks recently.Although spatial variations are explicitlyreflected in CSI measurements,the representation differences caused by small contextual changes are easilysubmerged in the fluctuations of multipath effects,especially in device-free Wi-Fi sensing.Most existing datasolutions cannot fully exploit the temporal,spatial,and frequency information carried by CSI,which results ininsufficient sensing resolution for indoor scenario changes.As a result,the well-liked machine learning(ML)-based CSI sensing models still struggling with stable performance.This paper formulates a time-frequency matrixon the premise of demonstrating that the CSI has low-rank potential and then proposes a distributed factorizationalgorithm to effectively separate the stable structured information and context fluctuations in the CSI matrix.Finally,a multidimensional tensor is generated by combining the time-frequency gradients of CSI,which containsrich and fine-grained real-time contextual information.Extensive evaluations and case studies highlight thesuperiority of the proposal.

关 键 词：Wi-Fi sensing device-free CSI low-rank matrix factorization 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]