On adaptive threshold intervals for stopping recursive least squares in the space of signals
Volume 7, Issue 1 (1996), pp. 27–38
Pub. online: 1 January 1996
Type: Research Article
Published
1 January 1996
1 January 1996
Abstract
In the papers (Kaminskas, 1973; Kaminskas and Nemura, 1975) the stopping rule of recursive least squares (RLS) is worked out using the length of the confidence interval for the respective current meaning of the true output signal of a linear dynamic system. The aim of the given paper is the development of techniques for calculating threshold intervals of respective criteria, used in such a stopping rule. In this connection adaptive threshold intervals based on the Cramer-Rao lower bound according to Pupeikis (1995) are proposed here, too. The results of numerical simulation by IBM PC/AT are given.