Optimal Estimation for Multiple Packet Dropouts Systems Based on Measurement Predictor

The stochastic packet dropouts often occur in the networked control systems (NCSs), which can degrade performance or stability of the systems. An optimal estimation of the dropped data during the transmissions is often inevitable. This paper has proposed a new compensation method for NCSs with packet dropouts using measurement predictor, the estimated measurement values for the lost information are obtained by using all the weighted known measurement values instead of the latest one. Based on the estimated measurement values, the optimal estimators including filter, predictor, and smoother are developed via an innovation analysis approach for multiple packet dropouts systems with white and colored measurement noises. Simulations show that the proposed estimators based on all the known information improve the estimation accuracy of the lost information, the precision of state estimation can also be improved accordingly.