A UNIFYING CONSTRUCTION OF ORTHONORMAL BASES FOR SYSTEM-IDENTIFICATION

This paper develops a general and very simple construction for complet e orthonormal bases for system identification. This construction provi des a unifying formulation of many previously studied orthonormal base s since the common FIR and recently popular Laguerre and two-parameter Kautz model structures are restrictive special cases of the construct ion presented here. However, in contrast to these special cases, the b asis vectors in the unifying construction of this paper can have arbit rary placement of pole position according to the prior information the user wishes to inject. Results characterizing the completeness of the bases and the accuracy properties of models estimated using the bases are provided.