Spatio-temporal spectral analysis by eigenvalue decomposition of a signal AOA matrix

Summary form only given, as follows. The estimation of the angles of arrival and the center frequencies of multiple narrowband sources is investigated. The simplest case, in which all the sources with a known common center frequency are located in the same plane as the sensor array, can be viewed as a one-dimensional (1-D) spectral problem. Generally speaking, however, the sources have unknown and different center frequencies, and the sources are not located in the same plane as the sensor array so that the angle of arrival (AOA) of an incident source is a vector, θ=(α,β). Thus, the problem of the simultaneous estimation of AOAs and center frequencies of multiple narrowband sources can be regarded as a three-dimensional (3-D) spectral estimation problem. A new high-resolution method for solving this 3-D problem is presented. The signal model is formulated for a rectangular sensor array with delay taps