Decomposition based least squares estimation algorithm for non-uniformly sampled multirate systems

In order to reduce the computational load of the recursive least squares (RLS) algorithm, a decomposition based least squares algorithm is developed for non-uniformly sampled multirate systems. The main ideal is to decompose the identification model of the non-uniformly sampled systems into several submodels with smaller dimensions and fewer parameters based on the hierarchical identification principle and then to use the least squares principle to estimate the parameters of each subsystem. The analysis and simulation results indicate that the proposed algorithm can give highly accurate estimates.