基于HHT人步行状态髋关节角度信号的分析方法

目的 提出一种基于希尔伯特-黄变换（Hilbert-Huang transform，HHT）分析人步行状态髋关节角度信号的方法 ，并验证其可行性。方法 首先，利用加速度传感器与陀螺仪组成的髋关节角度测量平台，测量健康人步行状态髋关节角度。其次，对此信号进行集合经验模态分解（ensemble empirical mode decomposition, EEMD），得到各本征模态函数（intrinsic mode functions, IMF），再对不同尺度的模态函数进行分析与组合。最后，对原信号进行Hilbert谱分析。结果 得到反映运动模式的特征信号以及髋关节旋转轨迹所表示的步态特征。Hilbert谱显示出主运动模式内的波内频率调制现象与步频特征。结论 此方法 适用于步态疾病患者的康复与治疗，可以有效地将髋关节角度信号不同频率尺度的特征信号进行分解，实现中心修正与滤波，达到自适应分析患者步态信号的目的 。

Method for analyzing the signal of hip joint angles during human walking using Hilbert-Huang transform

Objective To propose a method for analyzing the hip joint signals during human walking based on Hilbert-Huang transform (HHT) method and verify its feasibility. Methods First, the hip joint angles of one healthy subject were measured by using the hip joint measuring platform composed of acceleration sensors and gyroscopes. Second, all intrinsic mode functions (IMFs) at different scales, which could be further analyzed and combined, were obtained by applying the ensemble empirical mode decomposition (EEMD) to original signals. Finally, the Hilbert spectrum of original signals were plotted and analyzed. Results The signals representing different motion modes as well as gait characteristics indicated by rotating track of the hip joint were obtained. The Hilbert spectrum could show the intra-wave frequency modulation in the main motion mode and the characteristics of walking frequencies. Conclusions This method can be used in rehabilitation and treatment of patients with gait diseases. By using this method, the characteristic signals of the hip joints at different frequency scales can be effectively decomposed, and the post-processing signals can be filtered and centrally corrected, so as to adaptively analyze gait signals of the patients.