|
Activity Number:
|
544
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #309890 |
|
Title:
|
Using an Adaptive Control Filter To Predict the Synchronization of Time Series
|
|
Author(s):
|
Morris Morgan*+ and Carolyn Morgan
|
|
Companies:
|
Hampton University and Hampton University
|
|
Address:
|
Olin Engineering Building, Hampton, VA, 23668,
|
|
Keywords:
|
time series ; adaptive control ; chaos ; stochastic time series
|
|
Abstract:
|
Past research used statistical measures (e.g., m-statistic, Kendall 'tau' statistic and permutation entropies) to assess the degree of synchronization in time series where the underlying dynamic models were known a priori. Order-based statistics that are robust to additive noise were used to control nonlinear oscillators, couple delayed equations and analyze nonlinear time series such as electrocardiogram data. Graphical techniques that quickly characterize such systems were investigated. Current work is directed at using an adaptive control procedure that provides statistical-based parameter estimates. The approach is applicable for series where the underlying models are not known but must be coupled dynamically in the presence of stochastic noise. An array of traditional filters (e.g., Butterworth, Chebyshev and Haar wavelet) will be used to improve the efficiency of this approach.
|
