Title :
Probing signals for model reference identification
Author :
Yuan, Joseph S C ; Wonham, W.M.
Author_Institution :
SPAR Aerospace Products, Limited, Toronto, Canada
fDate :
8/1/1977 12:00:00 AM
Abstract :
This paper examines certain properties of probing signals employed in a model reference identification procedure which guarantee identification of the parameters of a linear multivariable system. The property of persistent excitation is examined with respect to a new class of (persistently spanning) signals. In particular, this class of signals includes many (periodic or almost periodic) signals which have previously been considered as effective probing signals, The results yield concrete guidelines useful in the practical design of probing signals.
Keywords :
Adaptive methods; Linear systems, time-invariant continuous-time; Lyapunov methods; Parameter identification; Adaptive systems; Concrete; Convergence; Learning systems; MIMO; Riccati equations; Signal design; Signal processing; Stability criteria; Symmetric matrices;
Journal_Title :
Automatic Control, IEEE Transactions on
DOI :
10.1109/TAC.1977.1101556