It is shown that the conventioanl linear prediction (LP) methods (including various versions of the Prony method) and the matrix pencil method for extracting poles of data sequences can be unified under a generalized approach called subspace linear prediction (SLP) approach. The conventional LP methods can be considered as high-order SLP methods, while the matrix pencil method is a first-order SLP method. We also discuss a special form of the matrix pencil method for oversampled data. It is observed that, for oversampled data, the noise sensitivity of the least-square Prony method can be significantly improved without using singular value decomposition or other subspace decomposition algorithms.
|Original language||English (US)|
|Title of host publication||Fourth Annu ASSP Workshop Spectrum Estim Model|
|Publisher||IEEE Computer Society|
|Number of pages||4|
|State||Published - 1988|
ASJC Scopus subject areas