System Identification using Slow and Irregular Output Samples

Identification of systems with very slowly and irregularly sampled output data is considered. While the input is sampled uniformly and frequently enough, the output sampling time is assumed very long, say, much longer than the dominant time constants. The output sampling may be irregular. Dual-rate system with slow output samples is a special case of the problem studied here. Model structures for the identification of the fast rate models will be proposed. Output error method is proposed and motivated for parameter estimation. Consistence of the estimate is established and covariance expression of the parameters is derived. Simulation studies will be used to illustrate the method. The developed identification method can also be used in other fields where measurement delay is a major problem.