Parametric estimation of ocean waves using distributed source modeling

There have been several methods for estimation of directional power spectral density (DPSD) of ocean waves. The roll and pitch buoy (RPB) was proposed by Longuit-Higgins in 1963. All the classical methods are only usable when the arriving waves from different directions are point uncorrelated sources. Otherwise their robustness is in doubt and these algorithms would fail. The authors present a solution which is based on distributed source (DS) theory. They examined the RPB sensors as the array and prove that the array is free of ambiguity for all frequencies. A generalization of the MUSIC algorithm as an example is applied for parametric estimation of ocean waves. Most of high resolution array processing techniques could be thus modified for parameter estimation of ocean DS waves. Two extremum cases, i.e. coherent distributed (CD) and incoherent distributed (ID) ocean waves with different distribution models have been considered. All models have two special parameters to be estimated: the mean wave direction (MWD) and the angular width (AW). The efficiency of the method is confirmed by simulation. In addition, the estimation of MWD is unbiased and the SNR effect on the estimation is negligible, when the number of snapshots is large.