Adaptive Lattice-Based RLS Algorithms
There are a large number of algorithms that solve the least-squares problem in a recursive form. In particular, the algorithms based on the lattice realization are very attractive because they allow modular implementation and require a reduced number of arithmetic operations (of order N) –. As a consequence, the lattice recursive least-squares (LRLS) algorithms are considered fast implementations of the RLS problem.
KeywordsPrediction Error Adaptive Filter Lattice Algorithm Lattice Realization Forward Prediction
Unable to display preview. Download preview PDF.
- 4.M. Bellanger, Adaptive Digital Filters and Signal Processing, Marcel Dekker Inc., New York, NY, 1987.Google Scholar
- 5.S. Haykin, Adaptive Filter Theory, Prentice Hall, Englewood Cliffs, NJ, 4th edition, 2002.Google Scholar
- 12.V. J. Mathews and Z. Xie, “Fixed-point error analysis of stochastic gradient adaptive lattice filters,” IEEE Trans, on Signal Processing, vol. 31, pp. 70–80, Jan. 1990.Google Scholar
- 13.M. Reed and B. Liu, “Analysis of simplified gradient adaptive lattice algorithms using power-of-two quantization,” Proc. 1990 IEEE Intern. Symp. on Circuits and Systems, New Orleans, LA, pp. 792–795, May 1990.Google Scholar