On-Line System Identification in Real Time Using Minicomputer

Abstract

<p>By making use of measurements of the input and output of an unknown system, the characterizing parameters are found by using matrix pseudoinverse method. For noise-corrupted measurement, least squares estimation of the parameters are obtained.</p> <p>Stochastic approximation is employed to improve the estimation.</p> <p>Both methods are tested on-line in real time using the PDP-11/45 minicomputer while the system is simulated on the TR-20 analog computer.</p>