This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: EE 504 Homework #1 Due: March 11, 2005 P.1: We have studied the estimation of r.v. x from r.v. y in the last lectures. In one of the examples, the estimator had the form b x = ay + b . The minimum MSE for this estimator is derived as J ∗ = ρ 2 x (1 − ρ 2 xy ). We have noted in class that it is possible to reach zero error for this estimator (estimation matches the true value) if the correlation coeﬃcient of x and y is ± 1. In this problem we examine the reverse argument. Show that if x and y have the correlation coeﬃcient of ± 1, x has to be in the form x = ay + b , matching the structure of the estimator. P.2: We have derived the optimal MSE estimator as ϕ ( x ) = E { y  x } , where y is the r.v. to be estimated, x is the observation. Assume that the observation x on unknown r.v. y can take four different values, x ∈ { 1 , 2 , 3 , 4 } . We denote MSE estimator which minimizes the error E { ( y − c ( x )) 2 } by c ( x ) = E { y  x } ....
View
Full
Document
This note was uploaded on 10/29/2009 for the course EE ee 504 taught by Professor Candan during the Spring '09 term at Middle East Technical University.
 Spring '09
 candan
 Signal Processing

Click to edit the document details