Model based estimation approach to the missing data problem

The problem of missing data segments is experienced frequently in measurement systems in industry and research. This paper presents the technique of model estimation based method to arrive at near-accurate estimates of the missed data. The significance of this technique lies in its general applicability to linear, nonlinear, time invariant and stochastic systems. The steps of model based estimation technique is explained along with other known methods of missing data estimation. The technique is further demonstrated using real time data from a nonlinear time invariant system and validated by experimental data.