State-dimension reduction and filtering for linear systems under communication constraints
通讯约束下的线性系统状态降维与估计

We investigate how to reduce the state dimensions when estimating the states of a linear dynamic system with channel communication power constraints.To meet the requirements on the dimension number and communication power constraints of the parallel channels,we adopt the structure of differential pulse code modulation(DPCM) to produce the innovation as the transmitted signal;and a new method of state-dimension reduction is derived under the minimum error entropy estimation(MEEE) criterion of filtering at receiver.Furthermore,the problem of state estimability of the stochastic system and the observability of the corresponding deterministic system are analyzed by using information theoretic method.Analysis and simulation results show that the estimation performance of Kalman filter is optimal under communication power constraint when this dimension reduction method is applied.