State Estimation For Several Classes Of Discrete Networked Systems Via The Coding-Decoding Scheme

In the networked environment,the state estimation of dynamic systems has gradually attracted the attention of researchers.In fact,in order to analyze and study the performance of a networked dynamic system,it is often necessary to know the original state information of the system.On the other hand,in actual systems,due to various phenomena,such as data packet loss,network congestion,intermittent sensor failures,etc.,signals often appear to be lost.Therefore,some new methods are proposed to solve the state estimation problem of discrete networked dynamic systems,and it is a research topic with both challenges and broad application prospects.Aiming at the networked system under the digital communication channel,in this paper,an encoding-decoding method is introduced to discuss the problem of state estimation.Specifically,considering several types of uncertain discrete nonlinear systems,the encoder and decoder are designed to facilitate the transmission of digital channels,and the state estimation problem is studied in a networked environment.The specific research content is as follows:Firstly,the state estimation problem is discussed for a class of nonlinear discrete-time uncertain networked systems.According to the measurement output,the encoder and decoder are designed to make the signal smoothly pass through the digital transmission channel.The constructed encoder encodes the observation data into a specific codeword,and the corresponding decoder decodes the received codeword.Using linear matrix inequality technology and Lyapunov theory,the expression form of the observer gain matrix is given.Furthermore,a specific sampling interval is selected,an improved unified quantization method is adopted,and the adequacy detection conditions are given including the size of the coded letter and the coding period,so that the estimation error system is asymptotically stable in a limited digital communication channel.Finally,a Matlab simulation example is used to verify the feasibility and effectiveness of the estimation algorithm based on the encoding-decoding scheme.Secondly,the state estimation problem is investigated for a class of discrete-time networked systems with packet loss.In order to characterize the measurement loss in front of the digital channel,a cluster of independent random variables are introduced that obey the Bernoulli distribution.According to Lyapunov theory and using linear matrix inequality technology,the expression of the observer gain is given.Then,in order to facilitate the transmission of digital signals,an encoder and a decoder are constructed separately,an appropriate encoding-decoding program is designed,a specific sampling interval is selected,an improved unified quantization method is used,and sufficient conditions for the estimation error system to be detectable are given.Considering the two situations of not serious data packet loss and serious data packet loss,simulation examples are used to verify the feasibility and effectiveness of the estimation scheme.Thirdly,a state estimation algorithm is studied for uncertain discrete-time networked systems with measurement loss and random nonlinearity.In order to characterize the phenomenon of measurement loss and random nonlinearity,two clusters of independent random variables that obey Bernoulli distribution and have known conditional probabilities are introduced to study the state estimation problem of the system.Considering the loss of measurement and random occurrence of nonlinearity,an elastic observer is designed,and sufficient conditions for the existence of the observer are given.The Lyapunov theory is used to solve a series of linear matrix inequalities and the expression form of the observer gain matrix is given.Based on the observations,the encoder and decoder are constructed,and the corresponding encoding-decoding program is designed.In order to facilitate signal coding,an auxiliary system is introduced,and then reverse decoding is performed to obtain the decoded value of the system,so that the estimated error after decoding becomes asymptotically stable.Finally,a simulation example is used to verify the effectiveness of the estimation algorithm.