Laurie T. Fialkowski
Michael D. Collins
Naval Res. Lab., Washington, DC 20375
A source in an uncertain environment may be localized using focalization, which involves a parameter space that includes both source and environmental parameters. A source buried in noise may be localized using noise-canceling techniques, which involve matching both signal and noise, or eigen-processing techniques, which are capable of extracting signals from noisy data. These techniques have been combined to solve localization problems involving a source buried in noise in an uncertain environment. Noise-canceling techniques requires knowledge of the noise. Since it may not be practical to model the noise as part of the parameter search, it is assumed that the noise covariance matrix is obtained from data taken when the source is not present (mismatch associated with temporal environmental variability is included in the simulations). Eigen-processing techniques are less effective than noise-canceling techniques at suppressing noise and require source motion when multiple sources are present. However, they have the advantage of not requiring a priori knowledge of the noise.