Title :
Direct estimation of multipath signals
Author :
Farrier, D.R. ; McLeod, F.N.
Author_Institution :
Southampton University, Southampton, UK
Abstract :
A new technique is proposed for the detection and estimation of multipath signals. The technique is based on using source subspace eigenvectors rather than the more frequently used noise subspace eigenvectors[1],[2]. The technique may be applied to linear and to curved arrays. The technique relies on a criterion used in Factor Analysis, called Varimax. The method uses this criterion to transform source eigenvectors into possible signal direction vectors. Comparisons with the MUSIC method [1] are made; results demonstrate that the new technique has better resolution than this method.
Keywords :
Array signal processing; Covariance matrix; Eigenvalues and eigenfunctions; Equations; Karhunen-Loeve transforms; Multiple signal classification; Oceans; Sensor arrays; Signal resolution; Vectors;
Conference_Titel :
Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '86.
DOI :
10.1109/ICASSP.1986.1168571