Filtering For Distributed Systems Based On Event Trigger In Uncertain Environment

Distributed computing is the sharing of information between multiple pieces of software and can run on one or more machines connected by a network of computers.Distributed computing divides a complex application into multiple parts and allocates them to multiple computers,thus reducing the running time and realizing resource sharing.It is necessary to exchange information between neighbor sensors in a distributed system,but frequent information exchange will lead to resource tension,network congestion,and eventually system instability,and even packet loss and transmission delay.Therefore,it is necessary to design an algorithm to reduce the frequency of information exchange between sensors.The principle of the event-triggered mechanism is to reduce the frequency of data transmission by the sensor.The trigger condition of the sampling time is set through the "event".When the trigger condition is met,the sensor transmits data,thus reducing the data transmission frequency and improving the resource utilization.However,how to use the filter method for efficient distributed processing becomes a challenge.According to the working principle of event-triggered and its role in the field of state estimation,the distributed system filtering method based on event-triggered becomes a hot research topic.The main work is summarized as follows:(1)In order to solve the system security problem,the privacy protected distributed discrete time varying linear system filtering method in wireless sensor networks is studied.The systems are vulnerable to network attacks when exchanging information between sensors in the network,as well as putting their security at risk.A system with privacy-preserving is designed for this purpose.A set-membership estimator is designed by adding privacy noise obeying the Laplace distribution to state at the initial moment.Simultaneously,the differential privacy of the system is analyzed.On this basis,the real state of the system and the existence form of the estimator for the desired distribution are analyzed.Finally,simulation examples are given,which prove that the model after adding differential privacy can obtain accurate estimates and ensure the security of the system state.(2)Aiming at the limitation of sensor resources,a distributed system filtering method with communication constraints is studied.In wireless sensor networks,the frequent information exchange between sensors depends on a large number of network communication resources.However,in practical applications,network communication resources are limited,and frequent data exchange between sensors will lead to network congestion,transmission delay or packet loss.In order to reduce the information transmission frequency between sensors and improve the utilization rate of communication resources,a set-membership estimator based on dynamic event-triggered with privacy-preservation is designed,and the privacy of the system is analyzed to ensure the security of real data in the system.On this basis,a recursive convex optimization method is used to analyze the event-triggered set-membership estimator with differential privacy method,and then the steady-state performance of the system is studied.Finally,an example is given to verify the feasibility of the proposed distributed filter containing the privacy protection method.(3)According to the communication delay problem,the distributed system filtering method with communication delay is studied.In practical applications,due to the influence of complex environment and many uncertain factors,the sensor will inevitably produce information collision or information retransmission in the process of information exchange,and it cannot transmit or receive information immediately.This leads to the delay of information transmission and system time lag,which makes the state estimation of the system complicated and even produces wrong results.In order to solve the problem of time delay,the observation sequence with time delay is reorganized by new information recombination analysis method,and the observation sequence without time delay and new information sequence are obtained.Based on the new observation sequence,the output measurement model without delay is established,and then a set-membership estimator with privacy-preservation is designed,and the stability of the estimator and the system is analyzed.Finally,the feasibility of introducing the privacy-preservation time-delay system model is verified by simulation experiments.