An RLS Algorithm for the Identification of Impulse Responses with Particular Symmetric Properties

In order to effectively solve system identification problems, it is always desirable to make use of particular characteristics of the impulse responses to be identified. In this paper, such an approach is illustrated by focusing on the estimation of systems with particular symmetric/antisymmetric features, which are handled using combinations of bilinear models. In the presented framework, we develop a recursive least-squares (RLS) algorithm customized for such models. When assessed against the conventional RLS counterpart, the proposed algorithm achieves significantly improved results, in terms of the employed performance measure. Simulation results validate the appealing performance of the designed algorithm.