有界噪声下具有一步时延和丢包的网络化估计

Networked Estimation with One-step Delay and Packet Dropout Under Bounded Noise

针对一类噪声统计特性未知情况下具有一步时延和数据丢包的网络化系统状态估计问题进行研究。首先,引入满足伯努利分布的随机变量描述一步时延和数据丢包现象,并基于此随机变量构建量测补偿模型。其次,通过量测补偿数据设计依概率信息的状态估计器,并借助于有界递归优化算法构造估计器均方误差上界,将求解估计器增益问题转化为求解凸优化问题,同时给出使估计器均方误差在每一时刻都有界的稳定性条件。最后,通过数值仿真对所提方法进行验证,仿真结果证明了所提方法的有效性。

The state estimation problem for a class of networked systems with one-step delay and data packet dropout in the presence of unknown noise statistical properties is studied.Firstly,a random variable satisfying the Bernoulli distribution is introduced to describe the one-step delay and data packet dropout,and a measurement compensation model is constructed based on this random variable.Secondly,a state estimator based on probability information is designed by the measurement compensation data,and the upper bound of the mean square error(MSE)of the estimator is constructed by means of a bounded recursive optimization algorithm,and the estimator gain can be determined by solving the transformed convex optimization problem.Meanwhile,the stability condition is given such that the MSE of the designed estimator is bounded at every step.Finally,the proposed method is verified by numerical simulation,and the simulation results prove the effectiveness of the proposed method.