Array modeling and signal detection for nonGaussian data
C.R. Baker · 2002
Representations are obtained for a stochastic process obtained as the output from an M-element array of sensors and due to the sum of received data and sensor noise. No assumptions are made on the statistical properties of the data, other than that the internal noise generated by each sensor is stationary and Gaussian. The characterization is in terms of functions that can be estimated from data. A general characterization is obtained, based on a property of the array filters that will obtain in many applications. Applications to signal detection are discussed.>