Distributed Weighted State and Unknown Input Fusion for Multi-Sensor Systems With Correlated Noises

This paper is concerned with the distributed fusion filtering algorithm weighted by matrices in the linear minimum variance sense for multi-sensor networks with correlated noises. The target plant is disturbed by the unknown inputs as well as correlated noises. The correlation among noises is firstly removed by introducing auxiliary factors, and the uniform quantization is taken into consideration by characterizing as an additive white noise obeying uniform distribution. For each sensor node, an upper bound for the filtering error covariance is obtained in the presence of unknown inputs and uniform quantization effects and minimized by designing the local filter parameters. Then, a distributed fusion recursive filtering algorithm is proposed with the help of the matrix weighting technique. Finally, an illustrative example is provided to demonstrate the effectiveness of the proposed results.