Earticle

In this paper, the fusion filtering problem has been investigated for multi-sensor time varying systems with fading measurements, the probabilistic fading phenomenon of which is described by statistical means and variances. A real-time unbiased fusion filtering algorithm has been proposed in the sequential fusion frame. The asynchronous fading measurements are firstly transformed into pseudo fading measurements of the fusion state in the current fusion interval. A novel noise estimation method is presented to deal with the correlations during pseudo measurement noises, which is strictly deduced in the sense of Linear minimum-mean-square error (LMMSE). On this basis, these fading measurements are handled sequentially in their arriving sequence without matrix augmenting. The finial computer simulation demonstrates the effectiveness of the proposed sequential fusion filtering method.