• home
• current students
• graduate
• events

Presenter: Hannan Lohrasbi
Supervisor: Dr. T. Aaron Gulliver

Date: Wed, August 13, 2014
Time: 09:00:00 - 10:00:00
Place: EOW 430

ABSTRACT

Abstract

Energy based source localization methods have received great interest due to their low cost and simple implementation. The total least squares estimator has been shown to be efficient when errors exist in the system parameters as it compensates for both observation matrix and data vector errors. As a consequence, it provides better performance than conventional LS-based methods. An efficient two stage estimator is presented based on the total least squares measurement analysis. In first stage the mean square error of the energy based measurement model is minimized, and in the second stage the nonconvex model is transformed to a convex model to estimate the unknown parameter vector. Performance results are presented which confirm the efficiency of the proposed method which achieves the Cramer-Rao bound.