Data-Reusing Adaptive Filtering Algorithms with Adaptive Error Constraint

We present a family of data-reusing and affine
projection algorithms. For identification of a noisy linear finite
impulse response channel, a partial knowledge of a channel,
especially noise, can be used to improve the performance of
the adaptive filter. Motivated by this fact, the proposed scheme
incorporates an estimate of a knowledge of noise. A constraint, called
the adaptive noise constraint, estimates an unknown information of
noise. By imposing this constraint on a cost function of data-reusing
and affine projection algorithms, a cost function based on the adaptive
noise constraint and Lagrange multiplier is defined. Minimizing the
new cost function leads to the adaptive noise constrained (ANC)
data-reusing and affine projection algorithms. Experimental results
comparing the proposed schemes to standard data-reusing and affine
projection algorithms clearly indicate their superior performance.

Authors:

• Young-Seok Choi

Keywords:

• Data-reusing
• affine projection algorithm
• error constraint
• system identification.

References:

[1] S. Haykin, Adaptive Filter Theory, 4th edition, Upper Saddle River, NJ:
Prentice Hall, 2002. [2] H.-C. Shin, W.-J. Song and A. H. Sayed, “Mean-square performance
of data-reusing algorithms,” IEEE Signal Processing Lett., vol. 12,
pp. 851–854, Dec. 2005.
[3] B. A. Schnaufer and W. K. Jenkins, “New data-reusing LMS algorithms
for improved convergence,” in Proc. Asilomar Conf., Pacific Groves,
CA, May 1993, pp. 1584–1588.
[4] K. Ozeki and T. Umeda, “An adaptive filtering algorithm using an
orthogonal projection to an affine subspace and its properties,” Electro.
Commun. Jpn., vol. 67-A, no. 5, pp. 19–27, 1984.
[5] H.-C. Shin and A. H. Sayed, “Mean-square performance of a family of
affine projection algorithms,” IEEE Trans. Signal Processing, vol. 52,
pp. 90–102, Jan. 2004.
[6] Y. Wei, S. B. Gelfand and J. V. krogmeier, “Noise-constrained least
mean square algorithm,” IEEE Trans. Signal Processing, vol. 49, No. 9,
pp. 1961–1970, Sep. 2001.
[7] S. Y. Choi, T.-W. Lee and D. S. Hong, “Adaptive error-constrained
method for LMS algotihms and applications,” Signal Processing, vol. 85,
pp. 1875–1897, Oct. 2005.
[8] H.-C. Shin, A. H. Sayed and W.-J. Song, “Variable step-size NLMS and
affine projection algorithms,” IEEE Signal Processing Lett., vol. 11,
pp. 132–135, Feb. 2004.