Optimal filtering for systems with bounded random measurement delays

The optimal filtering problem is studied for network control systems subject to random measurement delays without time stamps. With the given probability distribution of the delay, a new measurement model is proposed subject to one measurement only received once and more than one measurement can received at one time. The delayed measurements are written as multiple channel measurements that contain the same state information as the original measurement and each channel has single constant delay. The optimal filter is derived by adopting a re-organized innovation analysis approach, and the filter is given in terms of Riccati difference equations.