Testing the interpretation of the Fermi Galactic center excess in terms of unresolved point sources

We introduce a new method to analyse populations of near-threshold sources in gamma-ray data. The method is based on a wavelet transform of the gamma-ray sky, using Mexican hat wavelets. The statistics of peak significances in the wavelet transform provides information about the spatial distribution and luminosity function of near-threshold sources. We apply this method to gammaray data from the inner Galaxy, in order to test the millisecond pulsar interpretation of the Fermi Galactic center excess. We find evidence for the unresolved point sources causing the excess emission at the > 4σ level. For plausible luminosity functions, this new component explains 100% of the observed excess emission.