The Family of Fast Least Squares Algorithms for Adaptive Filtering

Three representative algorithms from the family of Fast Least Squares (FLS) algorithms are reviewed. They correspond to the transversal FIR filter, the lattice-ladder structure and the rotation approach respectively. They all solve the least squares problem recursively and are convenient for adaptive filtering.They lead to three different levels of computational complexity and robustness to round-off errors. Each of them also has special implementation aspects and provides a particular set of signal parameters.Overall, these three types of algorithms offer a wide and welcome choice to the adaptive filter designer.