Authors : Sasmita Hembram; Madhusmita Soren; Pankajini Naik; Purnima Sethy

Volume/Issue : Volume 11 - 2026, Issue 4 - April

Google Scholar : https://tinyurl.com/2hw9vm3v

Scribd : https://tinyurl.com/3j6cdxnp

DOI : https://doi.org/10.38124/ijisrt/26apr1707

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : System identification plays a crucial role in accurately modeling dynamic systems, especially in noisy environments. This paper presents a comparative performance analysis of three adaptive algorithms: Least Mean Square (LMS), Recursive Least Square (RLS), and a proposed Hybrid algorithm. The evaluation is conducted under different Signal-to-Noise Ratio (SNR) conditions (10 dB, 20 dB, and 30 dB) using key performance metrics such as Mean Square Error (MSE), convergence iterations, tracking error, and steady-state error. Simulation results demonstrate that the Hybrid algorithm consistently outperforms conventional LMS and RLS algorithms in terms of MSE and tracking accuracy across all SNR levels. Specifically, the Hybrid approach achieves an improvement of approximately 5.18% over LMS and 9.44% over RLS, indicating enhanced estimation accuracy and robustness. Although all algorithms converge within similar iterations, the Hybrid model provides better stability and reduced error variance. These findings suggest that combining adaptive filtering techniques can significantly improve system identification performance in noisy environments, making the Hybrid algorithm a promising approach for real-world signal processing applications.

References :

1. L. Ljung, System Identification: Theory for the User, 2nd ed. Prentice Hall, 1999.
2. S. Haykin, Adaptive Filter Theory, 5th ed. Pearson, 2014.
3. B. Widrow and S. D. Stearns, Adaptive Signal Processing. Prentice Hall, 1985.
4. A. H. Sayed, Fundamentals of Adaptive Filtering. Wiley, 2003.
5. B. Widrow et al., “Adaptive noise cancelling: Principles and applications,” Proc. IEEE, vol. 63, no. 12, pp. 1692–1716, 1975.
6. S. Haykin, “The LMS algorithm,” IEEE Signal Processing Magazine, vol. 30, no. 1, pp. 30–38, 2013.
7. M. H. Hayes, Statistical Digital Signal Processing and Modeling. Wiley, 1996.
8. R. Arablouei and D. P. Mandic, “A class of hybrid LMS-RLS adaptive filters,” IEEE Trans. Signal Processing, vol. 62, no. 18, pp. 4785–4797, 2014.
9. Y. Chen, Y. Gu, and A. O. Hero, “Regularized least-mean-square algorithms,” IEEE Trans. Signal Processing, vol. 57, no. 11, pp. 427–439, 2009.
10. J. Benesty, Y. Huang, and J. Chen, “A fast recursive algorithm for optimum sequential signal detection,” IEEE Trans. Signal Processing, vol. 54, no. 11, pp. 427–437, 2006.