Study on direction finding method using wave distribution function with Gaussian distribution model

Abstract

The analysis of plasma waves propagating through the geomagnetosphere provides clues for understanding the medium information in wave generation and propagation processes. It is particularly important to understand the propagation characteristics of the waves in direction finding. In this research, we developed a novel algorithm based on the wave distribution function which is a direction finding method for the electromagnetic waves that satisfies dispersion relations in plasmas, and evaluated the method by using simulated data. Since direction finding by the wave distribution function is a so-called inverse problem, some conventional model was assumed to find the optimum solution. In order to achieve more accurate estimation, in this paper, we report on an algorithm that introduces a discretization method and an integration method that fully utilizes the parameters and model features in the Gaussian distribution model, which is, in principle, the most effective model in the point that it has physically meaningful parameters. As the fitting preprocess for finding the solution, we propose an estimation method for approximate distribution by employing the concept of the energy function, and demonstrate its usefulness in discriminating waves having non-Gaussian distributions and in determining the initial parameters of the fitting by Gaussian distributions. © 2002 Wiley Periodicals, Inc. Electron. Comm. Jpn. Pt. 1.