Partitioned least squares

Recursive least squares (RLS) is undoubtedly the most widely used procedure in the field of recursive parameter estimation. For simply structured models, computation requirements do not cause too many problems in terms of time constraints. However, for higher order system models time aspects can become of importance. A method for reducing the complexity of recursive least squares whilst retaining the performance characteristics is therefore of interest. In this paper a novel partitioned least squares (PLS) algorithm is presented, in which estimates from several simple system models are combined by means of a Bayesian methodology of pooling partial knowledge. The method has the added advantage that, when the simple models are of a similar structure, it lends itself directly to parallel processing procedures, thereby speeding up the entire parameter estimation process by several factors.