Offers the first Bayesian approach to the source separation problemProvides all of the mathematical and statistical background needed, from statistical distributions and introductory Bayesian probability to prior hyperparameter assessment and estimation methodsIncludes MATLAB code for each model presented and makes the code available for download from the InternetCovers the multivariate regression model, the factor analysis model, the delayed source separation model, the dynamic mixing coefficient models, and the correlation model, all discussed from the Bayesian perspective Using a Bayesian approach, this book addresses the blind 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 and includes MATLAB code for each model discussed.