The tracking properties of a recursive parameter-bounding algorithm, referred to as the Dasgupta-Huang optimal bounding ellipsoid algorithm (DHOBE algorithm) are investigated. Conditions that ensure the existence of these 100% confidence regions in the face of small model-parameter variations are derived. For larger parameter variations, it is shown that the existence of 100% confidence regions is guaranteed asymptotically. A modification that enables the algorithm to track large variations in model parameter is proposed. Simulation results show that in general, the modified algorithm has a tracking performance comparable, and in some cases superior, to that of the exponentially weighted recursive-least-squares (RLS) algorithm.
展开▼
