Title :
Detection performance of the L1 beamformer in the presence of underwater burst noise
Author :
Barroso, Victor N. ; Moura, José F.
Author_Institution :
Dept. Eng. Electr. e Comput., Inst. Superior Tecnico, Lisboa, Portugal
Abstract :
The problem of detecting a directional random signal in independent non-Gaussian noise is addressed. In particular, a Gauss-Gauss mixture model is assumed for the total noise field. The approach consists of substituting a beamformer whose design is based on a least-absolute-value criterion for the MMSE (minimum mean square error) beamformer included in the optimum quadratic receiver for the Gaussian signal in the independent Gaussian noise problem. The analytical study and the reported simulation experiments show the robustness of the resulting suboptimum receiver in the presence of unexpected impulsive noise
Keywords :
random noise; signal detection; underwater sound; Gauss-Gauss mixture model; Gaussian noise; Gaussian signal; L1 beamformer; directional random signal; impulsive noise; least-absolute-value criterion; minimum mean square error; nonGaussian noise; optimum quadratic receiver; robustness; simulation experiments; suboptimum receiver; underwater burst noise; Analytical models; Detectors; Gaussian noise; Mean square error methods; Narrowband; Noise robustness; Sensor arrays; Signal design; Signal processing; Statistics; Underwater tracking; Vectors;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on
Conference_Location :
Albuquerque, NM
DOI :
10.1109/ICASSP.1990.116208
