Identification based on data with sampling delays

Most identification and design methods for discrete time data assume that the inputs and outputs are sampled at the same time instants. In practice, there is always a delay due to the conversion times in the analog to digital converters. In the paper it is shown how data with sampling delays can be used to obtain the model corresponding to synchronous sampling. The modification is simple and can be done after the estimation. It is only required that a suitable structure is used for the estimated model and that the delay is known