摘要

In wireless sensor networks, dependent quantized data are widely studied. We consider the problem of parameters estimation for dependent quantized communication data where some of the data are censored to the Fusion Center (FC). We propose a distributed imputation method to fill in the censored values, and introduce a two-step maximum likelihood estimation (MLE) method to estimate the unknown parameters of the sensor system. The lower bound of the variance of the estimator is discussed to show the asymptotic efficiency. Numerical example of the constant false alarm rate (CFAR) detection system demonstrate the effectiveness of the proposed scheme.