Title :
Exact confidence interval for magnitude-squared coherence estimates
Author :
Wang, Shouyan ; Tang, Mengxing
Author_Institution :
Lab. of Physiol., Univ. of Oxford, UK
fDate :
3/1/2004 12:00:00 AM
Abstract :
The magnitude-squared coherence function is widely used in many applications. The approximate confidence interval is only reliable for large data segments. In this letter, an iterative algorithm is provided to compute the exact confidence interval from the cumulative distribution function. In order to use the confidence interval conveniently in practice, some libraries are provided, using the iterative algorithm and cubic spline interpolation.
Keywords :
coherence; delay estimation; interpolation; iterative methods; signal detection; splines (mathematics); SNR estimation; confidence interval; cubic spline interpolation; cumulative distribution function; iterative algorithm; magnitude-squared coherence estimate; signal detection; signal-to-noise ratio; time delay estimation; Delay estimation; Distributed computing; Distribution functions; Interpolation; Iterative algorithms; Libraries; Linearity; Signal detection; Spline; Stochastic processes;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2003.822897
