Time-Frequency Feature Analysis of Human Motion Attitude in IMU-based Pedestrian Dead Reckoning

Jiang Yan; Yiming Zhang; Siting Zhou; Qikui Han

doi:10.6919/ICJE.202604_12(4).0046

Authors

Jiang Yan
Yiming Zhang
Siting Zhou
Qikui Han

DOI:

https://doi.org/10.6919/ICJE.202604_12(4).0046

Keywords:

Human Attitude Recognition; Pedestrian Dead Reckoning; Inertial Measurement Unit; Feature Extraction; Attitude Recognition; fTime-Frequency Domain Analysis.

Abstract

Aiming at the core bottlenecks restricting the improvement of positioning accuracy in existing Pedestrian Dead Reckoning (PDR) systems,including the separation between positioning parameter solution and human attitude recognition,the failure to establish a quantitative relationship between attitude features and positioning errors,and the imperfection of the PDR-oriented attitude feature system,this study carries out a time-frequency feature analysis of human motion attitude based on the Inertial Measurement Unit (IMU).In this research,three high-performance IMUs were used to collect triaxial acceleration and angular velocity data of six typical human motions with the sensor mounted on the chest.On this basis,sliding window filtering was determined as the optimal preprocessing scheme with its parameters optimized,a comprehensive time-frequency domain feature system was constructed,and the characteristics of the feature set were verified through multi-subject experiments.The results show that the proposed feature set can clearly distinguish different motion types and reveal the correlation law of time-frequency features of the six typical attitudes.This work provides technical support for high-precision PDR positioning,and has important reference value for the application of self-contained positioning with wearable inertial sensing technology.

Downloads

Download data is not yet available.

References

[1] YI Chunzhi,JIA Yicheng,JIANG Feng,and WANG Xiulai,“Inertial measurement unit based human motion intention recognition methods: Status and challenges,” CAAI Transactions on Intelligent Systems,vol.20,no.4,pp.763–775,2025.

[2] SUN Xiang,“Design and implementation of wearable human behavior recognition system based on IMU,” Master’s Thesis,Nanjing University of Information Science & Technology,Nanjing,Jiangsu,China,2017.

[3] SHEN Dehan and CHEN Changhao,“Pedestrian inertial navigation method based on time-frequency feature encoding neural network,” Navigation and Control,vol.24,no.6,pp.84–93,2025.

[4] WAN Pengbo,LI Xueqing,and TANG Yunqi,“Research on an improved pedestrian dead reckoning algorithm,” Electronic Measurement Technology,vol.47,no.11,pp.69–77,2024.

[5] DI Ke,LIU Jiaxin,DU Jiajia,REN Jie,ZHAO Weifeng,and LIU Yu,“Improved research on step length estimation model of PDR algorithm,” Journal of Electronic Measurement and Instrumentation,vol.36,no.11,pp.178–185,2022.

[6] WEI Xuanxuan,“Human attitude recognition algorithm based on inertial sensors,” Intelligent Computer and Applications,vol.12,no.6,pp.97–101,105,2022.

[7] ZHANG Xiaohong,LUO Kegan,TAO Xianlu,HU Xin,and LIU Wanke,“A multi-position PDR algorithm based on state recognition of wearable MEMS sensors,” Geomatics and Information Science of Wuhan University,vol.46,no.12,pp.1791–1801,2021.

[8] CHEN Jialong,“Human pose detection and implementation based on multi-modal information fusion,” Master’s Thesis,Heilongjiang University,Harbin,Heilongjiang,China,2021.

[9] LI Jinghui,“Research on human attitude recognition based on multi-sensor information fusion,” Master’s Thesis,Shandong University,Jinan,Shandong,China,2014.

[10] WANG De,“Research on human motion pattern and personal identity recognition algorithm based on IMU,” Master’s Thesis,Ningxia University,Yinchuan,Ningxia,China,2020.

[11] ZHANG Yunhao,HE Baiyue,YANG Xusheng,and ZHANG Wenan,“A review of human motion tracking methods based on wearable inertial sensors,” Acta Automatica Sinica,vol.45,no.8,pp.1439–1454,2019.

[12] D.Figo,P.C.Diniz,D.R.Ferreira,and J.M.Cardoso,“Preprocessing techniques for context recognition from accelerometer data,” Personal and Ubiquitous Computing,vol.14,no.7,pp.645–662,2010.

[13] YIN Xiaoling,CHEN Xiaojiang,XIA Qishou,et al.,“Human motion state recognition based on built-in sensors of smart phones,” Journal on Communications,vol.40,no.3,pp.157–169,2019.

[14] SUN Wei,SONG Ruyi,and DING Wei,“Adaptive discrimination of gait feature extraction based on K-means clustering,” Science of Surveying and Mapping,vol.44,no.12,pp.29–34,2019.

[15] CHEN Peng,LIAO Ming,LIU Wuping,et al.,“Analysis and application of key technologies for indoor positioning integrating trilateration and PDR,” Bulletin of Surveying and Mapping,no.2,pp.17–21,2019.

[16] XU Yongtao,“Research on human attitude recognition and trajectory reconstruction method based on inertial sensors,” Master’s Thesis,Southeast University,Nanjing,Jiangsu,China,2022.