scholarly journals A MISO-ARX-Based Method for Single-Trial Evoked Potential Extraction

2017 ◽  
Vol 2017 ◽  
pp. 1-10 ◽  
Author(s):  
Nannan Yu ◽  
Lingling Wu ◽  
Dexuan Zou ◽  
Ying Chen ◽  
Hanbing Lu
Keyword(s):  
Evoked Potential ◽  
Moving Average ◽  
Extraction Process ◽  
Autoregressive Moving Average ◽  
Single Trial ◽  
Model Driven ◽  
Single Output ◽  
Multiple Input ◽  
Novel Method ◽  
Electroencephalographic Activity

In this paper, we propose a novel method for solving the single-trial evoked potential (EP) estimation problem. In this method, the single-trial EP is considered as a complex containing many components, which may originate from different functional brain sites; these components can be distinguished according to their respective latencies and amplitudes and are extracted simultaneously by multiple-input single-output autoregressive modeling with exogenous input (MISO-ARX). The extraction process is performed in three stages: first, we use a reference EP as a template and decompose it into a set of components, which serve as subtemplates for the remaining steps. Then, a dictionary is constructed with these subtemplates, and EPs are preliminarily extracted by sparse coding in order to roughly estimate the latency of each component. Finally, the single-trial measurement is parametrically modeled by MISO-ARX while characterizing spontaneous electroencephalographic activity as an autoregression model driven by white noise and with each component of the EP modeled by autoregressive-moving-average filtering of the subtemplates. Once optimized, all components of the EP can be extracted. Compared with ARX, our method has greater tracking capabilities of specific components of the EP complex as each component is modeled individually in MISO-ARX. We provide exhaustive experimental results to show the effectiveness and feasibility of our method.

scholarly journals Development of Fuzzy Based Autoregressive Moving Average Exogenous Input Model for Water Flow in Nigerian Kanji Hydro-Power Dam

2021 ◽  
Vol 7 (2) ◽  
Author(s):  
D. O Araromi
Keyword(s):  
Controller Design ◽  
Moving Average ◽  
Model Development ◽  
Autoregressive Moving Average ◽  
Hydro Power ◽  
Model Driven ◽  
Robust Control System ◽  
Input Variables ◽  
Water Outflow ◽  
Safety Operation

Design of robust control system for any system requires model-driven approach. Therefore, it becomes imperative to develop a dynamic model suitable for controller design on safety operation of hydropower dam for power production in Kanji dam in Nigeria. Model for reservoir flow was developed in MATLAB environment using Fuzzy Based Autoregressive Moving Average Exogenous Input (FARMAX) model structure in this study. The data used for model development covered a period of ten years (2003-2013). It consists of water inflow (WI), water outflow (WO) and spillage (S). WI and S are input variables while WO was the output variable. The model obtained using the unsmoothed data with an outlier gave -14.115%, -0.302 and 610.317 for fit, R2 and RMSE, respectively. Unsmoothed data with no outlier gave -13.802%, -0.295 and 608.643 corresponding to fit, R2 and RMSE, respectively. The model obtained using the smoothed data in the presence of an outlier gave 80.533%, 0.962 and 104.113 for fit, R2 and RMSE, respectively. Smoothed data in the absence of outlier gave 81.533%, 0.962 and 99.637 for to fit, R2 and RMSE, respectively. FARMAX has the best fit value of 87.8774% when number of rules was equal to 3 with optima model order of 3 1 4 3. The model can serve as a decision support system in evaluating the optimal reservoir operation policies in real time.

scholarly journals Future Trend Forecast by Empirical Wavelet Transform and Autoregressive Moving Average

Sensors ◽  
2018 ◽  
Vol 18 (8) ◽  
pp. 2621 ◽  
Author(s):  
Qiusheng Wang ◽  
Haipeng Li ◽  
Jinyong Lin ◽  
Chunxia Zhang
Keyword(s):  
Complex System ◽  
Wavelet Transform ◽  
Moving Average ◽  
Parametric Model ◽  
Future Trend ◽  
Autoregressive Moving Average ◽  
System A ◽  
Empirical Wavelet Transform ◽  
Novel Method ◽  
System States

In engineering and technical fields, a large number of sensors are applied to monitor a complex system. A special class of signals are often captured by those sensors. Although they often have indirect or indistinct relationships among them, they simultaneously reflect the operating states of the whole system. Using these signals, the field engineers can evaluate the operational states, even predict future behaviors of the monitored system. A novel method of future operational trend forecast of a complex system is proposed in this paper. It is based on empirical wavelet transform (EWT) and autoregressive moving average (ARMA) techniques. Firstly, empirical wavelet transform is used to extract the significant mode from each recorded signal, which reflects one aspect of the operating system. Secondly, the system states are represented by the indicator function which are obtained from those normalized and weighted significant modes. Finally, the future trend is forecast by the parametric model of ARMA. The effectiveness and practicality of the proposed method are verified by a set of numerical experiments.

scholarly journals Clustering Input Signals Based Identification Algorithms for Two-Input Single-Output Models with Autoregressive Moving Average Noises

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-12
Author(s):  
Khalid Abd El Mageed Hag ElAmin
Keyword(s):  
Input Signal ◽  
Clustering Algorithm ◽  
Environmental Changes ◽  
Least Squares Method ◽  
Moving Average ◽  
Recursive Least Squares ◽  
Autoregressive Moving Average ◽  
Input Output ◽  
Single Output ◽  
Input Signals

This study focused on the identification problems of two-input single-output system with moving average noises based on unsupervised learning methods applied to the input signals. The input signal to the autoregressive moving average model is proposed to be arriving from a source with continuous technical and environmental changes as two separate featured input signals. These two input signals were grouped in a number of clusters using the K-means clustering algorithm. The clustered input signals were supplied to the model in an orderly fashion from cluster-1 up to cluster-K. To ensure that the output signal can be best predicted from the input signal which in turn leads to selecting good enough model for its intended use, the magnitude-squared coherence (MSC) measure is applied to the input/output signals in the cases of clustered and nonclustered inputs, which indicates best correlation coefficient when measured with clustered inputs. From collected input-output signals, we deduce a K-means clustering based recursive least squares method for estimating the parameter of autoregressive moving average system. The simulation results indicate that the suggested method is effective.

scholarly journals Bootstrap Order Determination for ARMA Models: A Comparison between Different Model Selection Criteria

2017 ◽  
Vol 2017 ◽  
pp. 1-12 ◽  
Author(s):  
Livio Fenga
Keyword(s):  
Selection Criteria ◽  
Moving Average ◽  
Information Criterion ◽  
Autoregressive Moving Average ◽  
Order Selection ◽  
Arma Models ◽  
Final Prediction Error ◽  
Novel Method ◽  
Order Determination ◽  
Selection Of

The present paper deals with the order selection of models of the class for autoregressive moving average. A novel method—previously designed to enhance the selection capabilities of the Akaike Information Criterion and successfully tested—is now extended to the other three popular selectors commonly used by both theoretical statisticians and practitioners. They are the final prediction error, the Bayesian information criterion, and the Hannan-Quinn information criterion which are employed in conjunction with a semiparametric bootstrap scheme of the type sieve.

Evoked Potential Enhancement Using a Neurophysiologically-based Model

2001 ◽  
Vol 40 (04) ◽  
pp. 338-345 ◽  
Author(s):  
A. Balaji Kavaipatti ◽  
O. Markusson ◽  
B. H. Jansen
Keyword(s):  
Evoked Potentials ◽  
Adaptive Filtering ◽  
Evoked Potential ◽  
Low Frequency ◽  
Model Parameters ◽  
Single Trial ◽  
Basic Procedure ◽  
Novel Method ◽  
Flashing Light ◽  
Electroencephalogram Eeg

Summary Objective: Single trial evoked potentials (EP) are generally obscured by the much larger spontaneous or background electroencephalogram (EEG). A novel method was developed to enhance single trial EPs. The potential of this approach was explored using actual flash evoked visual EPs. Method: The basic procedure is a variant of the adaptive filtering approach. At the core of our method is a mathematical, but neurophysiologically-realistic, nonlinear model of the cortical structures involved in generating EEG and EP activity. The model parameters are adjusted by a genetic algorithm in such a way that the model output resembles the actually observed pre-stimulus EEG activity. When post-stimulus EEG is passed through the inverse model, enhancement of the single trial EP should, theoretically, occur. Results: Evidence was found that, in case of visual evoked potentials obtained by flashing light through closed eyelids, alpha activity continues to around 150 ms post-stimulus, at which point a low frequency potential arises, cresting 100 ms later and disappearing after another 100 ms or so. Also, it was found that an individual’s response varies considerably from trial to trial. Conclusion: The inverse modeling approach presented here is effective at enhancing single trial EP activity. One potential application is to distinguish trials that contain a response from those that do not, which could result in improved ensemble averages.

scholarly journals Deterministic Chaos Detection and Simplicial Local Predictions Applied to Strawberry Production Time Series

Mathematics ◽  
2021 ◽  
Vol 9 (23) ◽  
pp. 3034
Author(s):  
Juan D. Borrero ◽  
Jesus Mariscal
Keyword(s):  
Time Series ◽  
Initial Conditions ◽  
Moving Average ◽  
Deterministic Chaos ◽  
Autoregressive Moving Average ◽  
Chaos Detection ◽  
Moving Average Models ◽  
Non Linear ◽  
Long Time ◽  
Novel Method

In this work, we attempted to find a non-linear dependency in the time series of strawberry production in Huelva (Spain) using a procedure based on metric tests measuring chaos. This study aims to develop a novel method for yield prediction. To do this, we study the system’s sensitivity to initial conditions (exponential growth of the errors) using the maximal Lyapunov exponent. To check the soundness of its computation on non-stationary and not excessively long time series, we employed the method of over-embedding, apart from repeating the computation with parts of the transformed time series. We determine the existence of deterministic chaos, and we conclude that non-linear techniques from chaos theory are better suited to describe the data than linear techniques such as the ARIMA (autoregressive integrated moving average) or SARIMA (seasonal autoregressive moving average) models. We proceed to predict short-term strawberry production using Lorenz’s Analog Method.

scholarly journals A Time-Varying MIMO Generalized Minimum Variance Controller for Servo Application

2013 ◽  
Vol 321-324 ◽  
pp. 1593-1596 ◽  
Author(s):  
Zheng Li ◽  
Guo Li Wang
Keyword(s):  
Performance Index ◽  
Moving Average ◽  
Minimum Variance ◽  
Autoregressive Moving Average ◽  
Time Varying ◽  
Average Model ◽  
Autoregressive Moving Average Model ◽  
Multiple Input ◽  
Control Objective ◽  
Moving Average Model

A generalized minimum variance controller is developed for multiple input and multiple output systems having time-varying dynamics. The plant to be controlled is described using a controlled autoregressive moving average model and the control objective is to minimize a generalized minimum variance performance index for servo applications.

scholarly journals Online NARMAX model for electron fluxes at GEO

2015 ◽  
Vol 33 (3) ◽  
pp. 405-411 ◽  
Author(s):  
R. J. Boynton ◽  
M. A. Balikhin ◽  
S. A. Billings
Keyword(s):  
Electron Flux ◽  
Moving Average ◽  
Autoregressive Moving Average ◽  
Earth Orbit ◽  
Output Data ◽  
Satellite Systems ◽  
Single Output ◽  
Average Electron ◽  
Solar Wind Parameters ◽  
Nonlinear Autoregressive

Abstract. Multi-input single-output (MISO) nonlinear autoregressive moving average with exogenous inputs (NARMAX) models have been derived to forecast the > 0.8 MeV and > 2 MeV electron fluxes at geostationary Earth orbit (GEO). The NARMAX algorithm is able to identify mathematical model for a wide class of nonlinear systems from input–output data. The models employ solar wind parameters as inputs to provide an estimate of the average electron flux for the following day, i.e. the 1-day forecast. The identified models are shown to provide a reliable forecast for both > 0.8 and > 2 MeV electron fluxes and are capable of providing real-time warnings of when the electron fluxes will be dangerously high for satellite systems. These models, named SNB3GEO > 0.8 and > 2 MeV electron flux models, have been implemented online at http://www.ssg.group.shef.ac.uk/USSW/UOSSW.html.

scholarly journals Data Filtering Based Recursive Least Squares Algorithm for Two-Input Single-Output Systems with Moving Average Noises

2014 ◽  
Vol 2014 ◽  
pp. 1-8 ◽  
Author(s):  
Xianling Lu ◽  
Wei Zhou ◽  
Wenlin Shi
Keyword(s):  
Least Squares ◽  
Least Squares Method ◽  
Moving Average ◽  
Recursive Least Squares ◽  
Noise Model ◽  
Autoregressive Moving Average ◽  
Data Filtering ◽  
Noise Transfer Function ◽  
Single Output ◽  
Recursive Least Squares Method

This paper studies identification problems of two-input single-output controlled autoregressive moving average systems by using an estimated noise transfer function to filter the input-output data. Through data filtering, we obtain two simple identification models, one containing the parameters of the system model and the other containing the parameters of the noise model. Furthermore, we deduce a data filtering based recursive least squares method for estimating the parameters of these two identification models, respectively, by replacing the unmeasurable variables in the information vectors with their estimates. The proposed algorithm has high computational efficiency because the dimensions of its covariance matrices become small. The simulation results indicate that the proposed algorithm is effective.

Sign in / Sign up

Export Citation Format

Share Document