Study On Estimation Problems For System With Packet Loss And Constraints

In recent years,the estimation of measurement packet dropping systems with constraints has attracted widespread attention from scholars at home and abroad in the fields of communication technology,positioning and navigation systems,and speech processing.First,for the measurement packet dropping system with constraints,we use two methods to solve the unbiased estimator.One solution is to derive the linear minimum mean square error estimator under unconstrained conditions,and then obtain the constrained estimator by projection method.The other method uses the time stamp technique and projection method to solve the estimator.In addition,for the measurement delay system with packet loss,we use the innovation reorganization theory to obtain the state estimator in the sense of quadratic filtering.The main research work of this thesis includes the following points:(1)For the measurement packet loss system with state constraints,we first solve the estimator under unconstrained conditions based on the linear minimum mean square error method.The estimator gain is obtained by solving a Riccati equation and a Lyapunov equation.Then,the projection theory is used to derive the constraint estimator on the basis of the unconstrained estimator.Because the above estimator needs to consider the convergence of Lyapunov equation in infinite time,the system matrix needs to be stable.In order to eliminate the limitation of the stability of the system matrix,another estimator is designed in this thesis by using the time-stamp technique,that is,the gain is calculated by solving a Riccati equation,and then the unconstrained estimate is directly projected onto the constrained surface to obtain constraint estimator.Finally,we use Matlab to design simulation experiments for the above two estimators respectively,and prove their effectiveness.(2)For measurement-delay systems with packet dropping,we use the innovation reorganization method to transform the time-delay system into a delay-free system,and then construct the delay-free system into a new enhanced system based on the quadratic filtering theory.Finally,two new Riccati equations and one Lyapunov equation are solved by the projection method for the new enhanced system,and the quadratic filtering algorithm can be obtained.In addition,we design simulation experiments to prove the effectiveness of the algorithm.