Abstract
Maps of the total electron content (TEC) of the ionosphere can be
reconstructed using data extracted from GPS signals. For historic and other
sparse data sets the reconstruction of TEC images is often performed using a
multivariate interpolation technique. Although there are many interpolation
methods available only a limited number, for example kriging, have been
applied to TEC data. This paper presents a quantitative comparison of various
commonly used algorithms for scattered data interpolation over a range of
sparsities. Techniques evaluated include a relatively new approach called
adaptive normalised convolution ANC, that has not previously been applied to
ionospheric reconstruction. The proposed evaluation scheme employs a
quantitative methodology applied to both simulated and real TEC data. Results
show that although the performance of kriging is good in many cases, it is
several times worse than the best performing techniques at some sparsities.
Natural neighbour interpolation has a better overall performance than kriging
for both simulated and TEC data. Although its performance is a few percent
worse than other methods for the simulated data, ANC produces the best
performance for the TEC reconstructions.
Original language | English |
---|---|
Pages (from-to) | 2153-2164 |
Number of pages | 12 |
Journal | IEEE Transactions on Geoscience and Remote Sensing |
Volume | 46 |
Issue number | 7 |
Early online date | 15 May 2008 |
DOIs | |
Publication status | Published - Jul 2008 |
Keywords
- Reconstruction
- Ionosphere
- TEC
- Interpolation