Fast implementation of sparse iterative covariance-based estimation for processing MST radar data

Abstract

Doppler estimation is an essential problem for the mesosphere–stratosphere–troposphere (MST) radar data for estimation of atmospheric parameters. The Doppler is estimated by computing the power spectral density using either parametric or non-parametric method. A recent class of spectral estimation technique referred as Sparse Iterative Covariance Based Estimation (SPICE) is introduced in literature. SPICE is a sound, user-parameter free, good resolution, iterative and globally convergent method which exhibits the enhanced results than the current spectral estimation methods, at the high computational cost. This letter presents the fast implementation of the SPICE algorithm for the MST radar data and the estimation of various wind parameters at a lesser computational time and is validated with the radiosonde data.