چكيده لاتين :
Fast Euclidean Direction Search (FEDS) [1] and Recursive Adaptive Matching Pursuit (RAMP) [2] are two recently introduced algorithms for adaptive filtering characterized by low computational complexity, good convergence, and numerical robustness. While conceived from quite different perspectives, we point out, that both algorithms are closely related and can be interpreted as different variants of 1) A matching pursuit procedure applied to a particular over-determined equation set, 2) A constrained Least Squares (LS) optimization problem, and 3) A Gauss-Seidel like iterative solution procedure applied to a normal equation. Both FEDS and RAMP have been demonstrated experimentally to possess excellent convergence behavior in several application scenarios. However, a tool for predicting the convergence of these algorithms based on second order statistics is lacking. This paper provides such a tool to study the convergence analysis of these adaptive filter algorithms. This tool relies on energy conservation arguments and doses not restrict to assume specific models for the regression data. Finally, we demonstrate through simulations that these results are useful in predicting adaptive filter performance.
