Partial-update NLMS algorithms with data-selective updating

In this paper, we present mean-squared convergence analysis for the partial-update normalized least-mean square (PU-NLMS) algorithm with closed-form expressions for the case of white input signals. The formulae presented here are more accurate than the ones found in the literature for the PU-NLMS algorithm. Thereafter, the ideas of the partial-update NLMS-type algorithms found in the literature are incorporated in the framework of set-membership filtering, from which data-selective NLMS-type algorithms with partial-update are derived. The new algorithms, referred to herein as the set-membership partial-update normalized least-mean square (SM-PU-NLMS) algorithms, combine the data-selective updating from set-membership filtering with the reduced computational complexity from partial updating. A thorough discussion of the SM-PU-NLMS algorithms follows, whereby we propose different update strategies and provide stability analysis and closed-form formulae for excess mean-squared error (MSE). Simulation results verify the analysis for the PU-NLMS algorithm and the good performance of the SM-PU-NLMS algorithms in terms of convergence speed, final misadjustment, and computational complexity.

[1]  Paulo S. R. Diniz,et al.  Adaptive Filtering: Algorithms and Practical Implementation , 1997 .

[2]  Yih-Fang Huang,et al.  Set-membership adaptive equalization and an updator-shared implementation for multiple channel communications systems , 1998, IEEE Trans. Signal Process..

[3]  S. Attallah,et al.  DCTLMS algorithm employing partial coefficient updates , 2000, Proceedings of the IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No.00EX373).

[4]  Andreas Antoniou,et al.  A new quasi-Newton adaptive filtering algorithm , 1997 .

[5]  Zakkula Govindarajulu.  Exact Lower Moments of Order Statistics in Samples from the Chi- Distribution (1 d.f.) , 1962 .

[6]  Alfred O. Hero,et al.  Stochastic partial update LMS algorithm for adaptive arrays , 2000, Proceedings of the 2000 IEEE Sensor Array and Multichannel Signal Processing Workshop. SAM 2000 (Cat. No.00EX410).

[7]  Richard J. Mammone.  Fast projection algorithms with application to voice echo cancellation , 1994 .

[8]  Scott C. Douglas,et al.  Adaptive filters employing partial updates , 1997 .

[9]  P. Diniz,et al.  Set-membership affine projection algorithm , 2001, IEEE Signal Processing Letters.

[10]  Mohammed A. Khasawneh,et al.  Analytical development of the MMAXNLMS algorithm , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[11]  Ronald L. Rivest,et al.  Introduction to Algorithms , 1990 .

[12]  Y. F. Huang,et al.  On the value of information in system identification - Bounded noise case , 1982, Autom..

[13]  Thomas Schertler,et al.  Selective block update of NLMS type algorithms , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[14]  Kutluyil Dogançay,et al.  Selective-partial-update NLMS and affine projection algorithms for acoustic echo cancellation , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[15]  H. Harter.  Expected values of normal order statistics , 1961 .

[16]  Tyseer Aboulnasr,et al.  Complexity reduction of the NLMS algorithm via selective coefficient update , 1999, IEEE Trans. Signal Process..

[17]  H. Godwin.  Some Low Moments of Order Statistics , 1949 .

[18]  Stefan Werner,et al.  Set-membership binormalized data-reusing LMS algorithms , 2003, IEEE Trans. Signal Process..

[19]  S. Douglas.  A family of normalized LMS algorithms , 1994, IEEE Signal Processing Letters.

[20]  Kutluyil Dogancay,et al.  Adaptive filtering algorithms with selective partial updates , 2001 .

[21]  Dirk T. M. Slock,et al.  On the convergence behavior of the LMS and the normalized LMS algorithms , 1993, IEEE Trans. Signal Process..

[22]  Shirish Nagaraj,et al.  Set-membership filtering and a set-membership normalized LMS algorithm with an adaptive step size , 1998, IEEE Signal Processing Letters.

[23]  S.C. Douglas,et al.  Analysis and implementation of the max-NLMS adaptive filter , 1995, Conference Record of The Twenty-Ninth Asilomar Conference on Signals, Systems and Computers.

[24]  Ioannis Pitas,et al.  Fast algorithms for running ordering and max/min calculation , 1989 .

[25]  Bruno O. Shubert,et al.  Random variables and stochastic processes , 1979 .

[26]  José Antonio Apolinário,et al.  Convergence analysis of the binormalized data-reusing LMS algorithm , 2000, IEEE Trans. Signal Process..