DocumentCode :
730535
Title :
Measure-transformed quasi maximum likelihood estimation with application to source localization
Author :
Todros, Koby ; Hero, Alfred O.
Author_Institution :
Ben-Gurion Univ. of the Negev, Beer-Sheva, Israel
fYear :
2015
fDate :
19-24 April 2015
Firstpage :
3462
Lastpage :
3466
Abstract :
In this paper, we consider the problem of estimating a deterministic vector parameter when the likelihood function is unknown or not expressible. We develop an estimator, called measure-transformed quasi maximum likelihood estimator (MT-QMLE), that minimizes the empirical Kullback-Leibler divergence between the transformed probability measure of the data and a hypothesized Gaussian probability distribution. By judicious choice of the transform we show that the proposed estimator can gain sensitivity to higher-order statistical information and resilience to outliers. Under some regularity conditions we show that the MT-QMLE is consistent, asymptotically normal and unbiased. Furthermore, we derive a necessary and sufficient condition for its asymptotic efficiency. The MT-QMLE is applied to source localization in a simulation example that illustrates its sensitivity to higher-order information and resilience to outliers.
Keywords :
Gaussian distribution; higher order statistics; maximum likelihood estimation; signal processing; vectors; MT-QMLE; asymptotic efficiency; deterministic vector parameter; empirical Kullback-Leibler divergence; higher-order statistical information; hypothesized Gaussian probability distribution; likelihood function; measure-transformed quasi maximum likelihood estimator; source localization; transformed probability measure; Covariance matrices; Maximum likelihood estimation; Probability; Robustness; Signal to noise ratio; Transforms; Higher-order statistics; parameter estimation; probability measure transform; robust estimation; source localization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on
Conference_Location :
South Brisbane, QLD
Type :
conf
DOI :
10.1109/ICASSP.2015.7178614
Filename :
7178614
Link To Document :
بازگشت