A Multiple-valued Decision-diagram-based Approach To Evaluate And Optimize The Reliability Of The Internet Of Things

As one of the hottest topics in the technological developments,the Internet of Things(IOT)has been widely used in various aspects of many industries,for example Smart Home,Intelligent Transportation,Food Safety and so on.A variety of multifunctional dynamic components have been appearing our daily life,more and more.Gradually,IOT is going to be the key position of the connection between human and social life.Its reliability will have a significant impact on our daily life,social production and economy development.The reliability of IOT system is one of the most important indicators to whether IOT system can work normally or not,in the specified working environment and working hours.IOT system can ensure the network connectivity and system stability,integrate the system services processes,fulfill the identification,transmission and analysis of the information,and finally actualize the intelligent management of any objects,whatever environment or situation it meets.The research of system reliability has a great significance to the design and implementation of IOT,which is usually a large and complex system.Dynamic fault trees(DFT)have been widely used for years because they can easily provide a concise representation of the dynamic failure behaviors of general non-repairable fault tolerant systems.When it comes to analysis the discrete multi-valued logical functions,Multiple-valued Decision Diagram(MDD)can be a representative data structure.Dynamic fault trees can play a very effective role on the evaluation of the reliability of complex systems for a long time,which have various dynamic failure modes.At the same time,in order to improve the efficiency of the system reliability analysis,the traditional modularization methods have been developed.While in reality,using the modularization methods can lead the state explosion problem and it can be very serious,whose components are relevant for the large dynamic sub-trees.For the large and complex system,a Multiple-Valued Decision-Diagram-based approach to solve Dynamic fault trees,can have a great significance of the system reliability analysis,especially the aspects in system design,implementation,maintenance,and optimization of IOT.For the reliability analysis of IOT,the main work of this paper is as follow:(1)From the architecture of IOT,we analyze the structure of every system parts.In consideration of the functions and structures of system components,a comprehensive evaluation system about the reliability of IOT is proposed.Then we consider the problems and their characters which is possible we can meet in the reliability analysis of IOT.Dynamic fault trees usually have great results in describing the failure modes of large fault-tolerant systems.In order to meet the increasingly complicated,large-scale IOT systems,we introduce a Multiple-Valued Decision-Diagram-based DFT analysis approach,to evaluate the system reliability of IOT.The method introduces that dynamic logistic gatesdescribe the influence relationship between components and systems,then it uses multiple-valued variables to encode the dynamic gates.We can portray the various dynamic and static behavior of the whole system.The single system MDD can describe the failure modes of systems simply and vividly.A practical case of Dynamic fault tree is introduced as an example to verify the effect of our method.After calculating of failure rate in the system MDD,we can evaluate its reliability.In consideration of the experiment result,the method can have a great practical value,while IOT systems usually are large-scale tolerant systems.(2)Considering every sub-systems and countless nodes of IOT,we introduce several classic common situations of the IOT system failures.In consideration of transmission paths and nodes of system network,we analyze the system failures which are based on the failures of communication and terminal components.Based on the Markov method,we calculate the probability when a component is in a certain state in system failure modes.As an example,a network computing system which has a high reliability,is analyzed by using the MDD-based approach to solve DFTs.A compact system MDD can be generated,which describes the system failure modes vividly.It can largely help relieve the state explosion problem of traditional modularization method,by combining the state-space method and events-combination method.For the system reliability research of increasingly complex and large-scale IOT,this method,it has a certain reference value in the design,management,maintenance and optimization of IOT systems.