A Maximum Likelihood-Based Minimum Mean Square Error Separation and Estimation of Stationary Gaussian Sources From Noisy Mixtures.

In the context of independent component analysis, noisy mixtures pose a dilemma regarding the desired objective. On one hand, a “maximally separating” solution, providing the minimal attainable interference-to-source-ratio (ISR), would often suffer from significant residual noise. On the other hand, optimal minimum mean square error (MMSE) estimation would yield estimates that are the “closest possible” to the true sources, often at the cost of compromised ISR. In this paper, we consider noisy mixtures of temporally diverse stationary Gaussian sources in a semi-blind scenario, which conveniently lends itself to either one of these objectives. We begin by deriving the maximum-likelihood (ML) estimates of the unknown (deterministic) parameters of the model: the mixing matrix and the (possibly different) noise variances in each sensor. We derive the likelihood equations for these parameters, as well as the corresponding Cramér-Rao lower bound, and propose an iterative solution for obtaining the ML estimates (MLEs). Based on these MLEs, the asymptotically optimal “maximally separating” solution can be readily obtained. However, we also present the ML-based MMSE estimate of the sources, alongside a frequency-domain-based computationally efficient scheme, exploiting their stationarity. We show that this estimate is asymptotically optimal and attains the (oracle) MMSE lower bound. Furthermore, for non-Gaussian signals, we show that this estimate serves as a quasi-ML-based linear MMSE (LMMSE) estimate, and attains the (oracle) LMMSE lower bound asymptotically. Empirical results of three simulation experiments are presented, corroborating our analytical derivations.