Turbulence Time Series Data Hole Filling using Karhunen-Loeve and ARIMA methods
Abstract
Measurements of optical turbulence time series data using unattended instruments over long time intervals inevitably lead to data drop-outs or degraded signals. We present a comparison of methods using both Principal Component Analysis, which is also known as the Karhunen-Loeve decomposition, and ARIMA that seek to correct for these event-induced and mechanically-induced signal drop-outs and degradations. We report on the quality of the correction by examining the Intrinsic Mode Functions generated by Empirical Mode Decomposition. The data studied are optical turbulence parameter time series from a commercial long path length optical anemometer/scintillometer, measured over several hundred metres in outdoor environments.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2007
- Accession Number
- ADA472169
Entities
People
- C. O. Font
- E. Oh
- G. C. Gilbreath
- H. Nazari
- J. L. Chang
Organizations
- University of Puerto Rico at Mayaguez