This preview shows pages 1–4. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Detection and Estimation (Spring, 2009) Linear Bayesian Estimators NCTU EE P.1 Linear Bayesian Estimators ~ Linear MMSE Estimator 9 The optimal Bayesian estimators are difficult to determine in closed form, and in practice too computationally intensive to implement. 9 We consider the class of all linear (actually affine) estimators of the form and choose the weighting coefficients a n ’s to minimize the Bayesian MSE Linear minimum mean square error (LMMSE) estimator Remark: a N is added to allow for nonzero means of x and θ . Let a = [a a 1 … a N1 ] T , we have Detection & Estimation (Spring, 2009) Linear Bayesian Estimators NCTU EE P.2 Remarks: 1. The LMMSE estimator is identical in form to the MMSE estimator for jointly Gaussian x and θ . 2. If the means of θ and x are zero, 3. All that is required to determine the LMMSE estimator are the first two moments of p( x , θ ). ~ Geometrical Interpretations Here we assume that θ and x are zero mean. If not, we can always define the zero mean random variables θ ’ = θ E( θ ) and x ’ = x – E( x ). Now, we wish to find the a n ’s so that minimizes Detection & Estimation (Spring, 2009) Linear Bayesian Estimators NCTU EE P.3 We define the inner product between two random variables x and y as...
View
Full
Document
This note was uploaded on 07/21/2009 for the course EE IEE5703 taught by Professor Shengjyhwang during the Spring '09 term at National Chiao Tung University.
 Spring '09
 ShengJyhWang

Click to edit the document details