Feasibility of source separation in frequency domain

We focus on the feasibility of the source separation in the frequency domain. First, it is linked with the convergence speed towards gaussianity of signals after L-point discrete Fourier Transform. We test here a distance to gaussianity thanks to the spectral kurtosis. We analyse the influence of L, of the duration of the source tricorrelations and of a non linear filtering. We mainly develop the case of QARMA processes. The second point consists in the reconstruction of the spectra of the estimated sources from the signals identified at each frequency bin. Indeed, the source associated to the ith identified signal is not necessarily the same from one frequency bin to another. The algorithm efficiency is then illustrated on QARMA processes, including the procedures of separation and reconstruction.