MULTIVARIATE BAYESIAN STATISTICS: MODELS FOR SOURCE SEPARATION AND SIGNAL UNMIXING

Using a Bayesian approach, this book addresses the source separation problem important in diverse applications from areas such as acoustics, genetics, portfolio allocation, and signal processing. It provides all the background needed, then examines the instantaneous constant mixing model where both the observed vectors and unobserved sources are independent over time but can be dependent within each vector. The author presents two distinct ways of estimating parameter for each model discussed.