Reliability Analysis And Optimization Of IoT Monitoring System

With the rapid development of big data,cloud computing,5G and other technologies,the global industry-internet of things have encountered a new round of historical opportunities for development.The application of it is being more and more extensive,covering all walks of life,from smart grid,intelligent transportation,intelligent logistics,intelligent agriculture,Internet of vehicles,smart home,smart medical care,to intelligent haze monitoring system.With the development of its technology and the popularization of its application,the reliability of its system becomes more and more prominent.As a result,how to ensure the reliability of its system has become a hot topic.This dissertation mainly focuses on the reliability analysis and optimization of its monitoring system.Considering the key pollution source monitoring system as background,taking the realization of the reliability of the perception layer in its monitoring system as the research goal,taking the sensing data as the basis,a reliable topology structure is designed,the fault-tolerant mechanism of the key nodes is studied,the data reliability inversion and correction model is proposed,and the reliability of its monitoring system is evaluated and applied.The dissertation is summarized as follows.(1)A modular node deployment scheme for internal and external validation is proposed.A reliable topology ensures a stable and dependable monitoring system.Considering the key pollution source monitoring system as background,a modular node deployment scheme for internal and external validation is proposed,related parameters and constraints are analyzed quantitatively,and the inherent rules between topological structure parameters is explored.Reliability is a critical index in Io T-based applications.Thus,the reliability of a multilevel cluster structure is presented using reliability block diagrams.For a reliable remote transmission model,different redundant systems are implemented.Furthermore,the reliability value and Mean Time To Failures(MTTF)are calculated.Simulation results show that the hexagonal cluster deployment scheme effectively reduced the cost of deployment and network maintenance.The reliability of a multilevel cluster structure decreased with the increases of the number of basic area,especially when the reliability of the sensor node is low.The best results using a triple parallel redundant system for a remote transmission system is obtained.The triple parallel redundant system not only prolongs MTTF,but also improves the reliability of the remote transmission systemof an Io T-based monitoring system.(2)A fault tolerance mechanism that combines CH static backup and dynamic timing monitoring(SBDTM)is proposed.In Io T applications,a wireless sensor network(WSN)is deployed to collect data.Clustered routing protocols that effectively maintain the energy consumed by sensor nodes(SNs)are usually employed in WSNs.Cluster heads(CHs)are important in this type of protocol.An effective fault tolerance mechanism for CHs in this system can guarantee reliable data acquisition.In order to optimize the reliability of data acquisition and energy consumption of Io T-based monitoring system,a fault tolerance mechanism that combines CH static backup and dynamic timing monitoring is proposed,a CH reliability model based on the Markov model is developed,and the minimum number of CHs necessary to satisfy the given reliability requirement is obtained.The data structures and fault-tolerant operations are described,and the energy consumption and the latency of the recovery of the SBDTM mechanism are quantitatively analyzed.The proposed SBDTM mechanism can effectively reduce the total energy consumed by all of the nodes in a cluster,the average energy consumed by each node in the cluster,and the latency of recovery.Several experiments are conducted to precisely evaluate the performance efficiency of the proposed SBDTM.The simulation results show that compared with LEACH and NCHG,SBDTM is reasonably efficient in reducing the total network energy consumption and packet loss rate and increasing the network lifetime and throughput.(3)A reliability inversion and correction model of Io T monitoring data under multi-attribute condition is proposed.Due to deliberate human interference and destruction,as well as the failure of the system itself,the key pollution sources monitoring system will be interrupted or the monitoring data will be abnormal.In order to solve this problem,an effective model based on the inversion and correction of the monitoring data of internal and surrounding pollution sources is provided based on the node deployment scheme for internal and external validation.An inversion index system is built and normalized to the standard evidence value beforehand.Moreover,relative entropy minimization model is used to assign weights to the indexes.Furthermore,the pollution levels are stored quantitatively.Reliability and accuracy of the results are ensured.Then,according to the total prior probability of the pollution source monitoring data,the prior probability of each sub-attribute,the joint probability distribution and its conditional probability that lead to the occurrence,the reliability inversion and correction of the monitoring data of the key pollution sources are realized.Finally,the proposed algorithm is applied on the practical application.The results are analyzed and presented.The proposed method shows improved precision and high accuracy in results.(4)The reliability of Io T-based monitoring system is comprehensively evaluated by using AHP-Fuzzy comprehensive evaluation method.An Io T monitoring system has various functional modules and levels,with different functional properties and influencing and evaluation factors.The complexity of the structure and operation mechanism of an Io T monitoring system,the restrictions of actual conditions,and the limitations of people's understanding may hinder the ability to measure or accurately quantify the function of some components or the true state of the entire system.Therefore,the comprehensive evaluation of the reliability of complex functional hierarchy systems is often equivalent to the performance evaluation of the system.Combining the advantages of traditional qualitative and quantitative analysis methods,then the reliability of Io T-based monitoring system is comprehensively evaluated by using the AHP-Fuzzy comprehensive evaluation method.The weighting coefficients of the indexes are calculated using AHP while the fuzzy comprehensive evaluation is used to realize the multi-layer comprehensive evaluation.Finally the reliability of the key pollution sources monitoring system of a certain thermal power plant in Hebei Province is evaluated by using the proposed evaluation model.Case study shows that the proposed calculation method and model can effectively evaluate the reliability of the Io T-based monitoring system.