Non-iterative methods to estimate the in-flight noise properties of CMB detectors

We present a new approach for statistical inference on noise properties of CMB anisotropy data. We consider a Maximum Likelihood parametric estimator to recover the full dependence structure of the noise process. We also consider a semiparametric procedure which is only sensitive to the low frequency behavior of the noise spectral density. Both approaches are statistically robust and computationally convenient in the case of long memory noise, even under nonstationary circumstances. We show that noise properties can be consistently derived by such procedures without resorting to currently used iterative noise-signal methods. More importantly, we show that optimal (GLS) CMB maps can be obtained from the observed timestream with the only knowledge of the noise memory parameter, the outcome of our estimators.