ABSTRACT
I. INTRODUCTION System identification is concerned with the determination of an ap propriate mathematical model which adequately describes the inputoutput characteristics of a system under consideration. The existing literature on the subject includes a large part which is based on treatment in discrete-time (DT) [1-4] and a considerable part which advocates and describes several approaches based on continuous-time (CT) for identifica tion of physical systems. An elaborate justification for the choice of CT models for physical systems and several methods of identification for a variety of CT model forms may be found in Unbehauen and Rao [5] and Sinha and Rao [6]. This chapter is concerned with identification of dynamic systems in terms of CT models in a particular form which is termed “generalized moving average” for its structural similarity with its discrete time (DT) counterpart. The term does not reflect the literal sense of moving average in this case.