The authors propose a two-stage predictor which is capable of predicting ARMA processes accurately with a reduced number of parameters at the expense of extra processing. This is achieved by cascading a classical linear predictor, where the constraint on the predictor order is relaxed, and a pole-zero recursivelike structure. A new optimum time-domain gradient algorithm for the linear prediction process is also presented. Computer simulations for both low- and high-order recursive filters show the excellent performance of the technique.
|Original language||English (US)|
|Title of host publication||Proceedings - IEEE International Symposium on Circuits and Systems|
|Number of pages||4|
|Publication status||Published - 1985|
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Electronic, Optical and Magnetic Materials