通讯访问约束条件下线性随机系统的状态可估计性

On state estimability of linear stochastic system under communication access constraints

本文从信息论的视角考察了线性离散随机系统在访问约束情况下的可估计性问题.针对系统中存在的访问约束所带来的通讯冲突现象,采用一个二值向量函数描述系统中各个输出信号的通讯状态.针对这类对象,根据由互信息定义的可估计性概念,讨论了访问约束系统的状态估计问题,得出了在输出检测信号传输存在访问约束情况下的状态可估计性条件,解析分析和仿真结果表明,存在合理的通信序列,使得系统在访问约束情况下,能够保持其状态可估计性.本文还证明了在访问约束条件下随机系统的可估计性和相应的确定性系统的能观性之间的等价关系.

The problem of estimability is discussed for linear multivariable discrete-time stochastic systems with com-munication access constraints.To deal with the communication constraints caused by the limited access,a binary-value function is used to describe the access status of the output signals.Based on the measure of mutual information,the problem of state estimability is discussed and a condition for estimability under access constraints is derived.Analytical analysis and simulation results show that there exist proper communication strategies which preserve the estimability of the system under communication access constraints.Furthermore,an equivalent relation between the estimability of stochastic system and the observability of corresponding deterministic system is proved under communication access constraints.