Correlated Measurement Fusion Steady-state Kalman Filtering Algorithms
Chenjian Ran, Hui Yu-Song, Lei Gu, Zili Deng · 2008
For the multisensor systems with correlated measurement noises and difierent measurement matrices, two correlated measurement fusion steady-state Kalman flltering algorithms are presented by using the weighted least squares (WLS) method. The principle is that a fused measurement equation is obtained by weighting the local measurement equations, and then it accompanies the state equation to realize the measurement fusion steady-state Kalman flltering. By using the information fllter, it is proved that they are functionally equivalent to the centralized fusion steady-state Kalman flltering algorithm, so that they have the asymptotic global optimality, and they can reduce the computational burden. They can be applied to the measurement fusion flltering and deconvolution for multichannel autoregressive moving average (ARMA) signals. Two numerical simulation examples verify their functional equivalence.