DocumentCode :
3012443
Title :
Application of spheroidal sequences to array processing
Author :
Forster, Philippe ; Vezzosi, Georges
Author_Institution :
Université de Rennes I, Rennes Cédex, France
Volume :
12
fYear :
1987
fDate :
31868
Firstpage :
2268
Lastpage :
2271
Abstract :
Eigenvector methods for narrow-band source location are based upon the key assumption that the spatial correlation matrix of the noise field is known. Unfortunately, these methods are very sensitive to errors upon it. This paper describes a method which performs a spatial filtering using discrete spheroidal sequences, before applying eigenvector methods. Thus, narrow spatial sectors are selected in which the noise field is easily modeled. In case of large arrays, the proposed method yields resolution properties far superior to the MUSIC technique with known noise correlation.
Keywords :
Acoustic sensors; Array signal processing; Eigenvalues and eigenfunctions; Filtering; Frequency; Multiple signal classification; Narrowband; Position measurement; Sensor arrays; Spatial resolution;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '87.
Type :
conf
DOI :
10.1109/ICASSP.1987.1169421
Filename :
1169421
Link To Document :
بازگشت