Bias propagation in the autocorrelation method of linear prediction
read more
Citations
Improving the Inertial Navigation System (INS) error model for INS and INS/DGPS applications
Automatic spectral analysis with time series models
Finite sample criteria for autoregressive order selection
Order selection for vector autoregressive models
References
Linear prediction: A tutorial review
Digital Processing of Speech Signals
Linear Prediction of Speech
Linear prediction of speech
Related Papers (5)
Frequently Asked Questions (12)
Q2. What causes the poor performance of the autocorrelation method?
The poor performance of the autocorrelation method is due to edge effects; incomplete terms in the residual cause a large bias in the residual variance, this bias is propagated via thedenominator of the Levinson–Durbin recursion and causes higher order reflection coefficients to be seriously biased as well.
Q3. What is the effect of a tapered data window?
A tapered data window decreases the edge effects, but increases the variance of estimated models, because effectively the number of data available for estimation is decreased.
Q4. Who was the associate editor coordinating the review of this manuscript?
The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Douglas D. O’Shaughnessy.
Q5. What is the effect of the bias on reflection coefficients?
the bias always tends to make the reflection coefficients smaller in absolute value, and this means that in general the poles will stay away further from the unit circle.
Q6. What is the bias in the autocorrelation coefficients?
1063–6676/97$10.00 1997 IEEEThe bias in the sample autocorrelation function (6) also causes bias in the estimated reflection coefficients.
Q7. What are the autocorrelation coefficients of the data?
The autocorrelation coefficients of the data are defined by(6)The autocorrelation coefficients are biased estimates of the theoretical autocorrelation coefficients of the process, because contains only nonzero products.
Q8. What is the expression for the bias in the reflection coefficients?
An expression for the bias in the reflection coefficients can be found by making a second order Taylor expansion and taking the expectation.
Q9. What is the definition of an autoregressive process of order?
An autoregressive process of order is described by a weighted sum of preceding signal values plus an independent identically distributed (i.i.d.) noise signal with variance(1)Manuscript received December 1, 1995; revised September 24, 1996.
Q10. What is the contribution of a rectangular window to the bias?
For a rectangular window, equals and since equals (the true first reflection coefficient of the process), this bias contribution is equal to .
Q11. What is the influence of the incomplete term on the second reflection coefficient?
The influence of the incomplete term is small because before minimization with respect to , all terms in the residual variance are of the same order of magnitude.
Q12. What is the effect of the biased residual variance?
The biased residual variance is propagated via the Levinson–Durbin algorithm (3) to the second reflection coefficient , and this causes to be considerably smaller than the true value .