China
Africa
Explore chapters and articles related to this topic
Case Studies
Published in Nicholas Stergiou, Nonlinear Analysis for Human Movement Variability, 2018
Anastasia Kyvelidou, Leslie M. Decker
One healthy young (age: 25 years; height: 179 cm; mass: 81.4 kg) and one healthy older adult (age: 84 years; height: 157 cm; mass: 66 kg) were screened and consented for participation in this study. COP was recorded for all participants using the Neurocom® Balance ManagerTM System (Neurocom International Inc., Clackamas, OR; 100 Hz). Each participant completed one trial of natural standing, each lasting 90 s. Subjects were instructed to stand with feet shoulder width apart and even weight distribution. The COP signal was analyzed in the AP and ML. The nonlinear dependent variable that we used was detrended fluctuation analysis (DFA, Chapter 7) to evaluate the temporal structure of the COP. DFA was used in the present case study due to the presence of possible trends introduced due to utilization of the Neurocom. DFA is a technique used to measure long-range correlations in nonstationary time series. Analyzing long-range correlations allows exploration of how movements on different time scales are related to each other. Stronger long-range correlations indicate that future movements rely more heavily on previous movements. Weaker long-range correlations indicate that movements rely less heavily on the memory of past events.
Introduction to ECG Time Series Variability Analysis: A Simple Overview
Published in Herbert F. Jelinek, David J. Cornforth, Ahsan H. Khandoker, ECG Time Series Variability Analysis, 2017
Herbert F. Jelinek, David J. Cornforth, Ahsan H. Khandoker
The variation in cardiac rhythm has mainly been suggested to be of nonlinear deterministic nature rather than due to stochastic noise. Nonlinear methods include a vast sample of biosignal processing algorithms. Examples are detrended fluctuation analysis (DFA), fractal dimension, symbolic dynamics, and entropy measures such as sample entropy, Renyi entropy, and the Lyapunov exponent. DFA is an estimate of the fractal correlation of the RR interval series; it provides an exponent expressing short-term correlations (alpha1) and another expressing long-term correlations (alpha2). Some of these measures are presented in other chapters of this book.
Modified Grey Wolf Randomized Optimization in Dementia Classification Using MRI Images
Published in IETE Journal of Research, 2022
N. Bharanidharan, R. Harikumar
The PCA technique is used to extract the small number of uncorrelated samples called principal components to represent a set containing a large number of uncorrelated samples. PCA technique is based on orthogonal transformation to find the principal components [15]. Usually, DFA is used to compute the statistical self-affinity of a signal in time series analysis. In DFA implementation, the variation from the trend is called fluctuation is computed initially and this fluctuation is measured over different window sizes [16]. HT is a linear operator that transforms the function of a real variable into another function of a real variable. HT incorporates the transformation in convolution with the Cauchy kernel [17,18]. K-means clustering is an iterative algorithm that splits the data points into K clusters based on the distance of data points from the mean [19].
Long memory and scaling behavior study of bulk freight rate volatility with structural breaks
Published in Transportation Letters, 2018
Xiaoxu Ding, Siyu Dai, Feier Chen, Yuqi Miao, Kang Tian, Yadong Zeng, Han Xu, Cao Qin
Four Hurst exponent estimators are prevalent in literature: the periodogram method, the R/S method, DFA, and the fit of the autocorrelation function (Lillo and Farmer 2004). Among these methods, DFA is applicable for both stationary and non-stationary time series and can give reasonable agreement both in real and in surrogate time series, thus it is widely employed by scholars (Kantelhardt et al. 2002). However, since DFA cannot be applied to multifractal spectrum of time series, the MF-DFA method, which is applicable to multifractal analyses, is proposed by Kantelhardt et al. on the basis of DFA. The MF-DFA method has its intrinsic advantage when employed to analyze the long-range correlation of time series, since it can effectively eliminate the trend component of each order in the series as well as detect the long-range correlation of signals superposed with noise and polynomial trend signals. In this paper, the MF-DFA method is employed to study the long memory feature of freight rate volatility and the robustness of long-memory inference on daily freight rate index prediction.
Application of multifractal detrended fluctuation analysis in fault diagnosis for a railway track circuit
Published in HKIE Transactions, 2018
Zicheng Wang, Yadong Zhang, Jin Guo, Lina Su
Many time-frequency methods have been applied to feature extraction of time series, including short time Fourier transform [7], wavelet transform [8] and empirical mode decomposition [9]. Nonetheless, the previous methods have their respective drawbacks which often produce unsatisfactory results in fault diagnosis application [10,11]. In 1994, a non-stationary signal processing method called detrended fluctuation analysis (DFA) was presented by Peng et al.[12,13]. DFA is a method for quantifying the scaling exponent of non-stationary time series. Through DFA, the unrelated trends of the non-stationary time series can be filtered out and the long-range correlations will be revealed. So classification can be performed according to the difference of the scaling exponents. However, DFA is only a monofractal analysis method and barely able to expose the underlying nonlinear dynamical mechanism in multifractal time series [14]. Multifractal detrended fluctuation analysis (MF-DFA) proposed by Kantelhardt et al. is an extension of the conventional DFA [15]. In MF-DFA, the fluctuation function in different order is used to analyse the scaling behaviour of time series at different levels. Consequently, the multifractality of non-stationary time series can be effectively revealed. MF-DFA has been successfully applied to data analysis in different scientific fields [16,17]. Therefore, MF-DFA was selected as a tool for analysing the locomotive signal amplitude envelope (LSAE) signals in this paper.