Scott A. Wright (Northrop Grumman Corporation), David Payne (Northrop Grumman Corporation), Russell Gillespie (Northrop Grumman Corporation)
Keywords: Space Weather
Abstract:
All electromagnetic radiation undergoes refraction as it propagates through the atmosphere. Tropospheric refraction is largely governed by interaction of the radiation with bounded electrons; ionospheric refraction is primarily governed by free electron interactions. The latter phenomenon is important for propagation and refraction of High Frequency (HF) through Extremely High Frequency (EHF) signals. The degree to which HF to EHF signals are bent is dependent upon the integrated refractive effect of the ionosphere: a result of the signals angle of incidence with the boundaries between adjacent ionospheric regions, the magnitude of change in electron density between two regions, as well as the frequency of the signal. In the case of HF signals, the ionosphere may bend the signal so much that it is directed back down towards the Earth, making over-the-horizon HF radio communication possible. Ionospheric refraction is a major challenge for space-based geolocation applications, where the ionosphere is typically the biggest contributor to geolocation error. Accurate geolocation requires an algorithm that accurately reflects the physical process of a signal transiting the ionosphere, and an accurate specification of the ionosphere at the time of the signal transit. Currently implemented solutions are limited by both the algorithm chosen to perform the ray trace and by the accuracy of the ionospheric data used in the calculations. This paper describes a technique for adapting a ray tracing algorithm to run on a General-Purpose Graphics Processing Unit (GPGPU or GPU), and using a physics-based model specifying the ionosphere at the time of signal transit. This technique allows simultaneous geolocation of significantly more signals than an equivalently priced Central Processing Unit (CPU) based system. Additionally, because this technique makes use of the most widely accepted numeric algorithm for ionospheric ray tracing and a timely physics-based model of the ionosphere instead of the typically used climatologically derived one, we assert that this technique improves geolocation accuracy.
Date of Conference: September 11-14, 2012
Track: Space Weather/Atmospherics