Analytical results on pseudo-polynomial functional-link neural units for blind density shaping

We deal with the problem of approximating the cumulative distribution of a quasi-stationary signal by means of a parametric function. We present an algorithm based on a quasi-polynomial adaptive transformation, and briefly recall an existing closely related technique, the 'mixture of densities' technique. Then through numerical simulations both on synthetic and real-world data we compare their performances in terms of convergence speed and computational complexity.