Title :
On the methods for solving Yule-Walker equations
Author :
Zhang, Hui-Min ; Duhamel, Pierre
Author_Institution :
CNET, Issy-les-Moulineaux, France
fDate :
12/1/1992 12:00:00 AM
Abstract :
The three well-known fast algorithms for the solution of Yule-Walker equations-the Levinson, Euclidean, and Berlekamp-Massey algorithms-are reviewed, and the relationship between each of them and the Pade approximation problem is shown. The algorithms are classified with reference to three criteria, namely: (1) the path they follow in the Pade table; (2) the organization of the computations (one-pass or two-pass); and (3) the auxiliary variables used. This classification shows that the set of known classical algorithms is not complete, and the missing variants are proposed. With these variants of the Berlekamp-Massey and Euclid algorithms, both forward and backward predictors can be obtained without additional cost. A unified representation of the two-pass algorithms is given in such a way that the application of the divide and conquer strategy becomes straightforward. A general doubling algorithm which represents all the associated doubling algorithms in an exhaustive way is provided
Keywords :
computational complexity; filtering and prediction theory; signal processing; Berlekamp-Massey algorithm; Euclidean algorithm; Levinson algorithm; Pade approximation problem; Yule-Walker equations; arithmetic complexity; backward predictors; divide and conquer strategy; doubling algorithm; fast algorithms; forward predictors; two-pass algorithms; Approximation algorithms; Arithmetic; Costs; Decoding; Digital signal processing; Equations; Error correction codes; Feedback; Reed-Solomon codes; Signal processing algorithms;
Journal_Title :
Signal Processing, IEEE Transactions on