Bayesian separation and recovery of convolutively mixed autoregressive sources

In this paper we address the problem of the separation and recovery of convolutively mixed autoregressive processes in a Bayesian framework. Solving this problem requires the ability to solve integration and/or optimization problems of complicated posterior distributions. We thus propose efficient stochastic algorithms based on Markov chain Monte Carlo (MCMC) methods. We present three algorithms. The first one is a classical Gibbs sampler that generates samples from the posterior distribution. The two other algorithms are stochastic optimization algorithms that allow to optimize either the marginal distribution of the sources, or the marginal distribution of the parameters of the sources and mixing filters, conditional upon the observation. Simulations are presented.

Nitrogen incorporation into tetrahedral hydrogenated amorphous carbon

Structural changes induced by the incorporation of nitrogen into ta-C : H films have been studied by Electron Energy Loss Spectroscopy, X-Ray Photoelectron Spectroscopy, Fourier Transformed Infrared Spectroscopy and Ultraviolet-Visible Spectroscopy. ta-C:H films have been synthesised using a low pressure Electron Cyclotron Wave Resonance (ECWR) source which provides a plasma beam with a high degree of ionisation and dissociation. Nitrogen was incorporated by adding N2 to the C2H2 plasma used for the deposition of ta-C : H films. The N/C atomic ratio in the films rises rapidly until the N2/C2H2 gas ratio reaches three, and then increases more gradually, while the deposition rate decreases steeply. Chemical sputtering of the forming films and the formation of molecular nitrogen within the films limit the maximum nitrogen content to about N/C = 0.6. For low nitrogen content the films retain their diamond-like properties, however as N/C atomic ratio increases, a polymeric-like material is formed, with >C=N- structures and terminating C=N and NH groups that decrease the connectivity of the network.