AN EFFICIENT APPROXIMATION TO THE QUADRATIC VOLTERRA FILTER AND ITS APPLICATION IN REAL-TIME LOUDSPEAKER LINEARIZATION

Nonlinear filtering based on the Volterra series expansion is a powerf ul and popular approach in signal processing. However, a serious probl em is the increased filter complexity as compared to linear filtering. This paper presents an efficient approximation to the second-order Vo lterra filter. The proposed filter structure, called multi-memory deco mposition (MMD), is composed of three linear FIR filters and one multi plier. Hence, the number of required filter operations is linear in th e filter memory length. MMD coefficient determination with respect to a second-order reference kernel is presented. Additionally, block-orie nted and adaptive algorithms are proposed which calculate the filter w eights from input and output measurements of an unknown system. The go od performance of the MMD model is demonstrated by simulations and in a real-time application. Therefore, the linearization scheme for the c ompensation of nonlinear distortions with a preprocessor is introduced . The preprocessor was implemented on a DSP system to reduce the nonli near distortions of electrodynamic loudspeakers in real-time. The resu lts show the gain in performance if an MMD filter is used instead of a Volterra filter. This is due to the fact that with a given computatio nal power longer memory lengths can be achieved by the MMD model.