Optimal joint azimuth-elevation and signal-array response estimation using parallel factor analysis

Abstract

We consider deterministic joint azimuth-elevation, signal, and array response estimation, and establish a direct link to parallel factor (PARAFAC) analysis, a tool with roots in linear algebra for multi-way arrays. This link affords a powerful identifiability result, plus the opportunity to tap on and extend the available expertise for fitting the PARAFAC model, to derive a deterministic (least squares) joint estimation algorithm, also applicable to multiple-parameter/multiple-invariance ESPRIT subspace fitting problems. These and other issues are demonstrated in pertinent simulation experiments.