State-space fractional-least mean square algorithm

In this work, a novel state-space adaptive algorithm, named state-space fractional-least mean square algorithm (SS-FLMS) is proposed. The proposed algorithm is based on the extension of fractional-least mean square (FLMS) algorithm for dynamic systems. The performance of the proposed algorithm is investigated for the estimation of a highly nonlinear system named Van der Pol oscillator and the outcomes are compared to the conventional state-space least mean square (SSLMS). The proposed algorithm outperforms the standard SSLMS algorithm by achieving fast convergence with lower steady-state. Simulation results are shown to validate the proposed claims.