Minimum variance linear estimation of amplitudes for exponential signal models

This correspondence presents the minimum variance consistent Linear estimat or for amplitude parameters in exponential signal model. A simple heuristic algorithm is presented to compute the weighting matrix that minimizes erro r variance; the resulting weighted least-squares estimator accounts fur the statistics of pole estimation errors. Additionally, analysis of binary dia gonal weighting matrices demonstrates that for unweighted least-squares, th e amplitude variance is reduced by truncating the Vandermonde system of equ ations.