On Supervised multilinear filtering: applications to system identification and antenna beamforming

Abstract

Linear filtering methods are well known and have been successfully applied to many engineering problems. However, they become unpractical when the parameter space is very large. The recently proposed assumption of system separability allows the development of computationally efficient alternatives to classical filtering methods in this scenario. In this work, we show that system separability calls for multilinear system representation and filtering. Based on this parallel, the proposed filtering framework consists of a multilinear extension of the classical Wiener-Hopf (WH) filter that exploits the separability property to solve the supervised multilinear filtering problem. System identification and antenna beamforming computer simulations were conducted to assess the performance of the proposed method. Our numerical results show our approach has smaller computational complexity and that it provides better estimation accuracy than the classical WH filter, which ignores the multilinear system structure.