A procedure for efficient estimation of multiple parameters in some nonlinear cases

This paper deals with a general procedure for asymptotically (in terms of signal-to-noise ratio) efficient estimation of mean parametrized Gaussian models. The proposed technique is based on an original theorem which is demonstrated. The resulting method allows in some nonlinear cases an efficient estimation without an iterative optimization algorithm (e.g. Gauss–Newton). The proposed framework is used to revisit and interpret an existing source localization method [Y.T. Chan, K.C. Ho, A simple and efficient estimator for hyperbolic location, IEEE Trans. Signal Process. 42(8) (August 1994) 1905–1915].