Spectral substraction and spectral estimation
Tipo de documentoTexto en actas de congreso
Fecha de publicación2000
Condiciones de accesoAcceso abierto
The problem of spectral subtraction, to estimate the parameters of a single source in colored noise, is used to show the relationships between the likelihood formulation and spectral density estimation. Reported previously as a filter bank processing for spectral estimation, it is shown that the normalized Capon estimate is the natural tool for source location in I-d and 2-d scenarios when the noise background estimate is faced as a spectral subtraction problem. Several simulations selecting I-d and 2-d apertures are used to show the degree of quality achieved with the proposed formulation. Also, the Periodogram test for incoherent detection is analyzed in front of the optimum test and the herein referred to as the Capon's test.
CitaciónLagunas, M.; Pérez, A. Spectral substraction and spectral estimation. A: Proc. X IEEE Workshop Statistical Signal and Array Processing. "Proc. X IEEE Workshop Statistical Signal and Array Processing". Pocono Mannor: 2000, p. 383-386.