lecture_24 - MIT OpenCourseWare http:/ocw.mit.edu 2.161...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
MIT OpenCourseWare http://ocw.mit.edu 2.161 Signal Processing: Continuous and Discrete Fall 2008 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms .
Background image of page 1

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

View Full DocumentRight Arrow Icon
F I R f i l t e r H ( z ) " d e s i r e d " o u t p u t { g } n { f } { e } e r r o r l e a s t - s q u a r e s d e s i g n a l g o r i t h m n n { d } n + - i n p u t o u t p u t { g } n f i l t e r c o e f f i c i e n t s b k } n { f } n Massachusetts Institute of Technology Department of Mechanical Engineering 2.161 Signal Processing - Continuous and Discrete Fall Term 2008 Lecture 24 1 Reading: Class Handout: MATLAB Examples of Least-Squares FIR Filter Design Proakis and Manolakis: Sec. 12.3 12.5 Stearns and Hush: Ch. 14 1 Least-Squares Filter Design We now look at a FIR filter design technique that is based on “experimental” data Given an input sequence { f n } , and a “desired” filtered output sequence { d n } , the task is to design a FIR filter M 1 H ( z ) = b k z k k =0 that will minimize the error { e n } = { d n }−{ g n } in some sense, where { g n } is the filter output. In particular, we will look at a filter design method that minimizes the mean-squared-error (MSE), where 2 MSE = E ± e n ² = E ± ( d n g n ) 2 ² 2 = E ± d 2 n ² + E ± g n ² 2 E { d n g n } 1 copyright c D.Rowell 2008 24–1
Background image of page 2
± ² ³ ² ³ ² ³ ² ³ 1 Massachusetts Institute of Technology Department of Mechanical Engineering 2.161 Signal Processing - Continuous and Discrete Fall Term 2008 Lecture 24 1 Reading: Class Handout: MATLAB Examples of Least-Squares FIR Filter Design Proakis and Manolakis: Sec. 12.3 12.5 Stearns and Hush: Ch. 14 Least-Squares Filter Design We now look at a FIR filter design technique that is based on “experimental” data " d e s i r e d " o u t p u t { d n } i n p u t F I R f i l t e r { g n } + e r r o r { f n } H ( z ) - { e n } o u t p u t { g n } b k f i l t e r c o e f f i c i e n t s { f } n l e a s t - s q u a r e s { e n } d e s i g n a l g o r i t h m Given an input sequence { f n } , and a “desired” filtered output sequence { d n } , the task is to design a FIR filter M 1 H ( z )= b k z k k =0 that will minimize the error { e n } = { d n }−{ g n } in some sense, where { g n } is the filter output. In particular, we will look at a filter design method that minimizes the mean-squared-error (MSE), where MSE = E e 2 n = E ( d n g n ) 2 = E d 2 + E g 2 2 E{ d n g n } n n 1 copyright ± c D.Rowell 2008 24–1
Background image of page 3

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

View Full DocumentRight Arrow Icon
±( ) ² ³ ³ ´ µ ³ ³ ³ ³ ³ ³ ³ ³ ³ and in terms of correlation functions MSE = φ dd (0) + φ gg (0) 2 φ dg (0) . For the FIR filter with coefficients b k , the output is { g n } = { f n }⊗{ b n } and from the input/output properties of linear systems (Lec. 22) φ ( n )= Z 1 H ( z ) H ( z 1 ) φ ff ( n ) M 1 M 1 = b m b n φ ( m n ) .
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/27/2012 for the course MECHANICAL 2.161 taught by Professor Derekrowell during the Fall '08 term at MIT.

Page1 / 14

lecture_24 - MIT OpenCourseWare http:/ocw.mit.edu 2.161...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online