ADAPTIVE LEAST SQUARES ALGORITHMS FOR TWO DIMENSIONAL SYSTEM IDENTIFICATION AND FILTERING: A UNIFIED APPROACH

摘要

In this paper, a unified approach is presented for adaptive Least Squares Two-Dimensional system identification and linear filtering. First, a unified deterministic Least Squares criterion is introduced, and subsequently utilized for the derivation of a general algorithmic framework for the adaptive Two-Dimensional system identification and filtering. Overdetermined, as well as underdetermined Least Squares Two-Dimensional adaptive algorithms are derived, from the proposed general adaptive scheme. In this way, known Two-Dimensional adaptive algorithms are interpreted as special cases of a general algorithmic form. Moreover, new adaptive algorithms are derived, following the proposed methodology.