Investigation on MNLT Method for 3-D Correlated Map Simulation of Sea-Surface Scattering

• Published in: IEEE geoscience and remote sensing letters 2018-10, Vol.15 (10), p.1595-1599
• Main Authors: Wei, Peng-Bo, Zhang, Min, Jiang, Wang-Qiang, Jiao, Yong-Chang
• Format: Article
• Language: English
• Subjects: Clutter; Computational modeling; Correlation; Data models; Data simulation; Estimation; Gaussian process; Interpolation; Memoryless nonlinear transform (MNLT) method; Methods; Predictive models; Radar; Radar clutter; Remote sensing; Scattering; sea scattering map; Simulation; Solid modeling; space-time correlation; Spectra; Stochastic processes; Stochasticity; Temperature (air-sea)
• ISSN: 1545-598X; 1558-0571
• DOI: 10.1109/LGRS.2018.2851373

In simulations of different kinds of radar clutter or scattering intensity distribution of terrain or sea surroundings, the memoryless nonlinear transform (MNLT) method is one of the most widely used approach for the simulation of non-Gaussian stochastic process or high-dimensional field, especially for the requirement of high-efficiency purpose. Concerning the MNLT method, the estimation of power spectral function usually exerts a strong impact on the quality of the implementation, which is not discussed in the previous literatures. This letter gives a study on the comparison of the 3-D correlated map simulations of sea-surface scattering using different ways of power spectral function estimation in the small to large sea scattering data simulation. From comparisons of the simulation results, it is observed that the interpolation-based expanding method is the optimal way to conduct power spectral function estimation, which can make a significant contribution to the scattering predictions and remote sensing simulations concerning the large maritime scenes.