Recursive Least Squares (RLS) Algorithm
The Recursive Least Squares (RLS) algorithm aims to minimize the weighted sum of squared errors for all input signals at each time instant, making it more adaptable to non-stationary signals. Unlike LMS algorithms that use ensemble averaging and produce consistent results under stable conditions, RLS employs time averaging where the optimal filter depends on the number of samples used for averaging. The algorithm implementation involves iterative updates of the inverse correlation matrix using the matrix inversion lemma, typically achieving faster convergence than LMS variants at the cost of higher computational complexity.