Optimal Filtering For Discrete Stochastic Systems With Time Delays And Missing Measurements

In this paper,the optimal filtering problem is investigated for a class of discrete stochastic systems with time delays and missing measurements.In the practical systems,it is the common problem for the systems with time delays and missing measurements.In order to estimate the state of systems accurately,it is great significance in theory and practice to research the systems with time delays and missing measurements.Firstly,the optimal filtering problem is investigated for a class of discrete stochastic time delays systems with multiplicative noises and random one-step sensor delay.The process noises and measurement noises of the systems are uncorrelated white noises.By using the innovative analysis approach and recursive projection formula,a linear optimal filter is designed such that the filtering error is minimized in the sense of mean square.The simulation example is given to illustrate the effectiveness and feasibility of the proposed filtering scheme.Then,the problem is extended to the general case.The optimal filtering problem is studied for a class of discrete stochastic time delay systems with randomly multiple sensor delays.In order to simplify the problem,the random multiplicative noises are not considered.Similarly,the optimal filter is designed for the addressed system.The simulation example is shown to illustrate the effectiveness and feasibility of the designed filtering.Secondly,the optimal filtering problem is researched for a class of discrete stochastic systems with multiplicative noises,finite-step autocorrelated process noises,random one-step sensor delay and missing measurements.The measurement noises of the systems are uncorrelated white noises.Based on the MMSE principle,a new linear optimal filter is designed such that the estimation error is minimized.The simulation example is shown to illustrate the effectiveness and feasibility of the proposed filtering scheme.Finally,the problem of consensus algorithm based on distributed unscented Kalman filtering is solved for a class of discrete stochastic nonlinear systems with missing measurements.The process noises and measurement noises of the systems are uncorrelated white noises.Based on the unscented transformation approach,a distributed unscented Kalman filtering is designed for a class of discrete stochastic nonlinear systems with missing measurements,and then,the problem is solved for the consensus algorithm based on distributed unscented Kalman filtering by using the consensus on information approach.The simulation example is shown to illustrate the uniformity of the proposed filtering scheme.