Self-tuning distributed fusion Kalman filter with asymptotic global optimality

For the multisensor systems with unknown model parameters and noise variances, by the system identification method, the estimators of the model parameters and noise variances can be obtained, and then substituting them into the steady-state optimal distributed fusion Kalman filter under the information filter form, a self-tuning distributed fusion Kalman filter is presented. Using the dynamic error system analysis (DESA) method, it is proved that the self-tuning distributed fusion Kalman filter converges to the steady-state optimal distributed fusion Kalman filter, so that it has asymptotic global optimality. It can be applied to the signal processing to obtain the self-tuning distributed fusion signal filter. A simulation example of a self-tuning fused filter for AR signal with 3-sensor shows its effectiveness.