Title :
On some nonparametric methods for detection of the number of signals
Author :
Yin, Y.Q. ; Krishnaiah, P.R.
Author_Institution :
University of Arizona, Tucson, AZ
fDate :
11/1/1987 12:00:00 AM
Abstract :
A simpler, more elegant, and more flexible method for estimating the number of signals is proposed. Strong consistency is proved. Bounds for error probabilities are given. This method depends on choosing a function with a parameter r. The previous methods correspond to the case r = 1. It seems the larger the r, the faster the convergence. We can choose r as any positive integer, even ∞. Applications are discussed, including the coherent cases.
Keywords :
Contracts; Convergence; Covariance matrix; Eigenvalues and eigenfunctions; Error probability; Gaussian noise; Signal processing; Sun; Vectors; White noise;
Journal_Title :
Acoustics, Speech and Signal Processing, IEEE Transactions on
DOI :
10.1109/TASSP.1987.1165063