Sliding Z Transform: Applications to convolutive blind source separation

TL;DR

The paper introduces the Sliding Z Transform (SZT), a novel signal processing tool inspired by the sliding DFT, which effectively aids in convolutive blind source separation by directly recovering sources from mixtures.

Contribution

It proposes a new definition of SZT and demonstrates its application to blind source separation, showing its simplicity and wide applicability.

Findings

Successfully recovers time-domain sources from convolutive mixtures

Demonstrates performance with various sliding window sizes

Applicable with robust linear blind separation algorithms like JADE

Abstract

The Z Transform is a mathematical operation in signal processing, which gives a tractable way to solve linear, constant-coefficient difference equations. Based on the classical Z transform and inspired by the thought of sliding DFT, a new definition of Sliding Z Transform(SZT) is introduced and deduced. Then this method is applied to blind source separation, four simulation results are presented to demonstrate its performance when the sliding window WIN is set. It can directly recover time-domain sources from the convolutive mixtures with the help of robust linear mixed blind separation algorithms(such as JADE) . It has simple principle and good transplantation capability and can be widely applied in various fields of digital signal processing.