Affine projection LMS adaptive algorithm with variable smoothing of weight update matrix

Document Type : Original Article

Authors

Department of Electrical and Computer Engineering, Qom University of Technology, Qom, Iran,

Abstract
Improving the convergence speed of adaptive filters is crucial for enhancing performance in
applications involving highly correlated input signals. In this paper, we propose a novel method to improve
the convergence performance of the affine projection LMS (AP-LMS) algorithm by incorporating a variable
smoothing approach for the weight update matrix. The smoothing parameter is dynamically assigned
based on the difference between the instantaneous and smoothed values of the weight update matrix.
Simulation results for FIR system modeling demonstrate that the proposed algorithm achieves superior
convergence performance in estimating system coefficients compared to competing adaptive algorithms for both stationary and non-stationary input signals.

Keywords

Subjects


[1] Ahmad, M. S., Kukrer, O., Hocanin, A. (2011). Recursive inverse adaptive filtering algorithm. Digital Signal Processing, 21(4), 491–496.
[2] Bekrani, M., Lotfizad, M., Khong, A. W. H. (2010). An efficient quasi-LMS/Newton adaptive algorithm for stereophonic acoustic echo cancellation. In Proc. IEEE Asia Pac. Conf. Circuits Syst. (pp. 684–687).
[3] Bekrani, M., Khong, A. W. H., Lotfizad, M. (2011). A linear neural network based approach to stereophonic acoustic echo cancellation. IEEE Transactions on Audio, Speech, and Language Processing, 19(6), 1743–1753.
[4] Bekrani, M., Bibak, R., Lotfizad, M. (2019). Improved clipped affine projection adaptive algorithm. IET Signal Processing, 13(1), 103–111.
[5] Douglas, S. (1995). The fast affine projection algorithm for active noise control. In Asilomar Conference on Signals, Systems and Computers (pp. 1245–1249).
[6] Farhang-Boroujeny, B. (2013). Adaptive filters: Theory and applications (2nd ed.). John Wiley and Sons.
[7] Gay, S. L. (1993). A fast converging, low complexity adaptive filtering algorithm. In IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (pp. 1245–1249).
[8] Gonzalez, A., Ferrer, M., Albu, F., Diego, M. (2012). Affine projection algorithms: Evolution to smart and fast algorithms and applications. In Proc. 20th European Signal Processing Conference (pp. 1965–1969).
[9] Haykin, S. (2014). Adaptive filter theory (5th ed.). Pearson. https://www.pearson.com/
[10] Hou, Y., Li, G., Zhang, H., Zhang, H., Zhao, J. (2024). Affine projection algorithm with outlier detection for robust filtering. IEEE Transactions on Circuits and Systems II: Express Briefs.
[11] Huang, F., Zhang, J., Zhang, S. (2016). Combined-step-size affine projection sign algorithm for robust adaptive filtering in impulsive interference environments. IEEE Transactions on Circuits and Systems, 63(5), 493–497.
[12] Husøy, J. H., Abadi, M. S. E. (2017). On the convergence speed of the normalized subband adaptive filter: Some new insights and interpretations. In International Symposium on Signals, Circuits and Systems.
[13] Koike, S. I. (2016). Analysis of affine projection normalized correlation algorithm. In International Symposium on Intelligent Signal Processing and Communication Systems.
[14] Lee, K., Bæk, Y., Park, Y. (2015). Nonlinear acoustic echo cancellation using a nonlinear postprocessor with a linearly constrained affine projection algorithm. IEEE Transactions on Circuits and Systems II: Express Briefs, 62(9), 881–885.
[15] Liu, D., Zhao, H. (2023). Affine projection sign subband adaptive filter algorithm with unbiased estimation under system identification. IEEE Transactions on Circuits and Systems II: Express Briefs, 70 (3), 1209–1213.
[16] Nita, V. A., Dobre, R. A., Ciochina, S., Paleologu, C. (2017). Improved convergence model of the affine projection algorithm for system identification. In International Symposium on Signals, Circuits and Systems.
[17] Petraglia, M. R., Haddad, D. B., Marques, E. L. (2016). Affine projection subband adaptive filter with low computational complexity. IEEE Transactions on Circuits and Systems II: Express Briefs, 63(10), 989–993.
[18] Poletti, M. A., Teal, P. D. (2021). A superfast Toeplitz matrix inversion method for single- and multi-channel inverse filters and its application to room equalization. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29(6), 3144–3157.
[19] Salman, M. S., Kukrer, O., Hocanin, A. (2012). A fast implementation of quasi-Newton LMS algorithm using FFT. In Proc. Inf. Commun. Technol. Appl. (pp. 510–513).
[20] Salman, M. S., Kukrer, O., Hocanin, A. (2020). A fast quasi-Newton adaptive algorithm based on approximate inversion of the autocorrelation matrix. IEEE Access, 8, 47877–47887.
[21] Tsinos, C. G., Diniz, P. S. R. (2019). Data-selective LMS-Newton and LMS-Quasi-Newton algorithms. In Proc. IEEE Int. Conf. Acoust. Speech Signal Process. (pp. 4848–4852).
[22] Wu, M., Xiong, N., Vasilakos, A. V., Leung, V. C. M., Chen, C. L. P. (2022). RNN-K: A reinforced Newton method for consensus-based distributed optimization and control over multiagent systems. IEEE Transactions on Cybernetics, 52(5), 4012–4026.
[23] Yuan, W. Y., Zhou, Y., Huang, Z. Y., Liu, H. Q. (2015). A VAD-based switch fast LMS/Newton algorithm for acoustic echo cancellation. In Proc. IEEE Int. Conf. Digit. Signal Process. (pp. 967–970).
[24] Zhao, H., Zheng, Z. (2016). Bias-compensated affine-projection-like algorithms with noisy input. Electronics Letters, 52(9), 712–714.
[25] Zhao, J., Ni, X., Li, Q., Tang, L., Zhang, H. (2024). Evolving order based affine projection sign algorithm for enhanced adaptive filtering. IEEE Signal Processing Letters, 31, 1530–1534.
[26] Zorkun, A. E., Salas-Natera, M. A., Martínez Rodriguez-Osorio, R. (2022). Improved iterative inverse matrix approximation algorithm for zero forcing precoding in large antenna arrays. IEEE Access, 10, 100964–100975. 
Volume 6, Issue 1
Winter 2025
Pages 89-103

  • Receive Date 11 October 2024
  • Revise Date 05 January 2025
  • Accept Date 06 January 2025