Continuous-time AR process parameter estimation in presence of additive white noise

Most existing work so far on continuous-time AR (CAR) parameter estimation concentrates on the noiseless measurement case. When measurement noise is p resent, our previous results on CAR parameter estimation need to he revised accordingly. Here we model the additive measurement noise as continuous-ti me white noise, and consider some approaches including average sampling and the direct LS method which we developed previously. Their advantages and d isadvantages in this application are discussed.