Statistical distribution of GPS amplitude scintillations

This study presents complementary cumulative distribution function (CCDF) as a statistical distribution apparatus for fitting GPS scintillations data. Three years of data at three levels of solar activity, 2002 (high), 2004 (moderate) and 2008 (low) from an equatorial anomaly crest station; Bogota (4.4 degrees N, 74.1 degrees W, dip 16.0 degrees N) [Colombia] in the West Coast of South America were used for the investigation. These data were grouped into daily, monthly and seasonal sets at three levels of solar activity, and tests were introduced on them to reject data from non-ionospheric sources of scintillation, such as multipath. Before fitting each set of data on a CCDF, the data were first visualized with the aid of scatter plots whereby the distributions exhibit non-Gaussian behavior. As a case study, at S4=0.3 threshold, during the year 2002 and 2004, the months of May-July showed probability of occurrence of the order of 0.01 (1% of the observed samples of a given set) for each month, while in 2008, this trend persisted to August (May-August, probability of occurrence of 0.01 or less (1%)). The tails of January and March's distributions were observed to be the heaviest at S4=0.3, although, a relatively heavy tail at this threshold was also observed during the month of November in the year 2004, and during March Equinox and December Solstice in terms of seasons. The heaviness of the tail at this threshold relaxes as solar activity decreases. The calculated probability of occurrences and those derived from the statistical distribution scheme show good consistency. The results presented in this study may be of assistance for future modeling and simulation studies. (C) 2011 Elsevier Ltd. All rights reserved.