A Class of Diffusion Proportionate Subband Adaptive Filters for Sparse System Identification over Distributed Networks

This paper aims to extend the proportionate adaptation concept design of a class diffusion normalized subband adaptive filter (DNSAF) algorithms. leads four extensions algorithm associated with different step-size variations, namely (DPNSAF), $$\mu $$ -law PNSAF (DMPNSAF), improved (DIPNSAF) and IPNSAF (DIIPNSAF). Subsequently, steady-state performance, stability conditions computational complexity proposed algorithms are investigated. For each extension performance has been evaluated using both real simulated data, where outcomes demonstrate accuracy theoretical expressions effectiveness