부산대학교 도서관

Highlights •The parameter estimation of Wiener model under time-varying delay, multi-rate sampling and irregular interference (impulse noise, outliers) is studied for the first time. Under the expectation–maximization framework, the Scale outlier model is introduced to model the contaminated output data. The time-varying delay at each sampling instant is assume to obey the uniform distribution. The proposed method can estimate the model parameter and time varying delay simultaneously.•For the presence of non-zero mean noise, the Location outlier model is introduced to model the contaminated output data. The proposed method can further estimate the bias of non-zero noise.•The probability graph model of the Wiener process is constructed. The idea of an auxiliary model is applied to acquire the unmeasured middle variable.•The identification convergence of the proposed algorithm is analyzed.