IDENTIFICATION OF CONTINUOUS-TIME SISO SYSTEMS VIA MARKOV PARAMETER-ESTIMATION

The paper presents a new approach towards unbiased estimation of the t ransfer functions of continuous-time SISO systems via Markov parameter s. A new set of parameters known as Markov-Poisson parameters is intro duced to generalise the definition of Markov parameters, thereby lendi ng much flexibility to the estimated model. An effective algorithm giv ing continuous-time transfer functions through the estimation of such models, and its recursive version are discussed in detail. Numerical e xamples are included to illustrate the proposed method and its superio rity over some existing ones.