The sequential implementation of array processors when there is directional uncertainty
Abstract
This report takes a global approach to the processing of information from an array of sensors. Specifically, the detection problem is considered. The general form of the likelihood ratio is derived based upon observables consisting of the Fourier coefficients of the observed random processes. For a stationary noise field consisting of a component independent from sensor to sensor and an additive directional component, the covariance properties of these Fourier coefficients are pursued as a function of the observation period length. Four canonical implementations of the optimal array processor structure are discussed: (1) one shot, (2) pseudo estimator, (3) two step, and (4) sequential. Of particular interest are three specific problems involving either signal or noise source location uncertainty.
- Publication:
-
Ph.D. Thesis
- Pub Date:
- August 1975
- Bibcode:
- 1975PhDT........80H
- Keywords:
-
- Fourier Transformation;
- Random Processes;
- Signal Detection;
- Signal Processing;
- Bayes Theorem;
- Estimators;
- Power Spectra;
- Probability Density Functions;
- White Noise;
- Electronics and Electrical Engineering