Long-Term Ionospheric Forecasting System
Abstract
Report developed under Small Business Innovative Research (SBIR) contract Topic AFO3-O16. The objective of this effort is to investigate the feasibility of an end-to-end global long-term (up to 3 days) ionospheric forecast model based on a fusion of several diverse technologies and to research the related probability density function (PDF) propagation formalism to characterize the forecast quality. We describe development of the proposed practical system based on a synthesis of several different technologies: (1) an artificial intelligence algorithm known as Support Vector Machines for predicting changes in solar wind from time sequences of solar images; (2) an empirical model of the high-latitude electric field potentials; (3) a physics-based ionospheric model coupled with efficient Kalman filter for forecasting the final ionospheric parameters of interest; and (4) a prototype error propagation scheme based on ensemble filters for computing evolution of forecast probability density functions.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 30, 2004
- Accession Number
- ADA427610
Entities
People
- Boris Khattalov
- Jason Boisvert
- Michael T. Murphy
- Tim Fuller-rowell