Error Analysis of Soil Roughness Parameters Estimated from Measured Surface Profile Data

Abstract

In experimental studies of electromagnetic wave scattering by rough surfaces, estimation of rough surface parameters, such as root-mean-square (rms) height and correlation length, from measured surface height-profile data is often required. For accurate estimation of these parameters, a data sample with sufficiently long record length is desirable. However, the criterion of the data length required for the estimation is not clear. In this research, we statistically estimate errors arising from the data length based on the interval estimates and check the results through a Monte Carlo simulation.