lecture_11

# lecture_11 - 2.160 System Identification, Estimation, and...

This preview shows pages 1–3. Sign up to view the full content.

2.160 System Identification, Estimation, and Learning Lecture Notes No. March 15, 2006 6.5 Times-Series Data Compression y ( t ) ) b 3 b 2 b 1 FIR u( t t Finite Impulse Response Model Consider a FIR Model ( ( ( ( t y ) = t u b 1) + t u b 2) + ⋅⋅ ⋅ + b t u m ) 1 2 m The transfer functions of (11) are then: G ( q ) = q B ), H ( q ) = 1 ( One-Step-Ahead Predictor 1 ( ı( 1 ) ( t y θ ) = H ( q ) G ( t u q ) + [ 1 H ( q ) ] t y ) ) ( = G ( t u q ) T t y ) = ϕ ( t ) : linear regression Given input data { t u 1 ), Nt } , the least square estimate of the parameter vector ( was obtained as N ı LS 1 ( 1 N = min arg 1 ( t y ) t y | ) ) 2 = ( R ( N ) ) f ( N ) (40) N t = 1 2 where N N T 1 ) ( R ( N ) = 1 ( t ) ( t ) = ΦΦ T and f ( N ) = ( t y t ) (41) N t = 1 N t = 1 Pro’s and Con’s of FIR Modeling Pros. LS ( LSE gives a consistent estimate N = 0 as long as the input sequence { t u ) } is lim ı N →∞ uncorrelated with the noise e(t) , which may be colored. Cons Two failure scenarios The impulse response has a slow decaying mode 1 11

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

View Full Document
The sampling rate is high The number of parameters, m, is too large to estimate. The persistently exciting condition rank = Φ rank full can hardly be satisfied. Check the eigenvalues of ΦΦ T (or the singular values of Φ ) λ 2 1 m It is likely n + 1 0 = = 1 2 n m Often m becomes more than 50 and it is difficult to obtain such an input series having 50 non-zero singular values. Time series data compression is an effective method for coping with this difficulty. Before formulating the above least square estimate problem, data are processed so that the information contained in the series of regressor may be represented in compact, low-dimensional form.
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 2.160 taught by Professor Harryasada during the Spring '06 term at MIT.

### Page1 / 8

lecture_11 - 2.160 System Identification, Estimation, and...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online