The spectral analysis of measured short waves can efficiently be carried out by the fast Fourier transform technique. Even though many present techniques can be used for the simulation of time series waves, these may not provide accurate and computationally efficient results for specific applications. Therefore, an alternate method is described here to provide spectral relationships with manimum loss of information. The measured short waves are expressed by the parametric models namely, authregressive or auto-regressive moving average algorithms which are the polynomical functions. Further reduced order autoregressive moving average spectral estimations are successfully applied to time series waves with different sampling rates. This paper emphasises the importance of selecting a suitable sampling interval for better estimates of parametric modelling and also for better statistical representation. Implementation of the above algorithms in a structural monitoring system has the potential advantage of storing more representations of wave environment than is currently possible because of the reduced size of the data sets.