Title: Adaptive recursive least squares method for parameter estimation of autoregressive models

Authors: Shazia Javed; Ghida Nazir; Nazir Ahmad Chaudhry; Ali Akgül; Muhammad Farhan Tabassum

Addresses: Department of Mathematics, Lahore College for Women University, 54000, Lahore, Pakistan ' Department of Mathematics, Lahore College for Women University, 54000, Lahore, Pakistan ' Department of Mathematics, Lahore College for Women University, 54000, Lahore, Pakistan ' Department of Mathematics, Art and Science Faculty, Siirt University, 56100 Siirt, Turkey; Department of Electronics and Communication Engineering, Saveetha School of Engineering, SIMATS, Chennai, Tamil Nadu, India ' Department of Mathematics, Lahore College for Women University, 54000, Lahore, Pakistan

Abstract: The recursive least squares (RLS) methods are extremely used to find the solutions of problems in many areas, such as communication, signal processing, optimisation and control. In this paper the RLS algorithm is modified for parameter estimation of regression models, such as the pseudo-linear ARMA (PS-ARMA) model and output error autoregressive (OEAR) model. The adaptive filtering technique with random input-output is used in the proposed recursive parameter estimation (RPE) algorithm to recursively predict the exact set of parameters for any regression model. The proposed method works by predicting an output signal that is adaptively improved to approximate the desired filter output. The experimental results are provided to prove the effectiveness of the proposed method.

Keywords: adaptive filter; autoregressive model; parameter estimation; random signals.

DOI: 10.1504/IJANS.2023.133733

International Journal of Applied Nonlinear Science, 2023 Vol.4 No.1, pp.72 - 89

Received: 08 Jun 2022
Accepted: 13 Dec 2022
Published online: 02 Oct 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article