{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

630_rec_9_sol

# 630_rec_9_sol - ENEE630 ADSP 1 RECITATION 9 w solution...

This preview shows pages 1–2. Sign up to view the full content.

ENEE630 ADSP RECITATION 9 w/ solution November 12/13, 2009 1. [Problem 8.3 continued] Assume v ( n ) and w ( n ) are white Gaussian random processes with zero mean and variance 1. The two filters in Fig. R9.1 are G ( z ) = 1 1 - 0 . 4 z - 1 and H ( z ) = 2 1 - 0 . 5 z - 1 . Figure R9.1: (a) Design a 1-order Wiener filter such that the desired output is u ( n ). What is the MSE? (b) Design a 2-order Wiener filter. What is the MSE? Solution: (a) R x = " r u (0) + 1 r u (1) r u (1) r u (0) + 1 # , and p xd = " r u (0) r u (1) # . The filter is w = R - 1 x p with MSE r u (0) - p H xd R - 1 x p xd . (b) Similar to (a), except R x = r u (0) + 1 r u (1) r u (2) r u (1) r u (0) + 1 r u (1) r u (2) r u (1) r u (0) + 1 , and p xd = r u (0) r u (1) r u (2) . MSE is still the same expression, i.e. r u (0) - p H xd R - 1 x p xd . Note: 1. In general, for a p -order AR model, given { σ 2 v , a 1 , a 2 , . . . , a p } , we can find { r (0) , r (1) , r (2) , . . . } ; and vice versa. They are related by Yule-Walker Equations. 2. r ( - k ) = r * ( k ) in general (and hence matrix R is Hermitian), and r ( - k ) = r ( k ) for real-valued signals. r (0) is the power of sequence u ( n ), and hence r (0) > 0 from physical point of view.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}