Abnormal Behavior Analysis And Pinning Control Of Networked Software

The development of the software industry has become the pillar industry of national economy.The popularity of Internet technology caused major changes in the software industry,which makes the software engineering into a new era of network software?In today's perplexing network environment,network software for their own behavior,structure and complex combination,which makes the new software system of safety and reliability of some trusted attribute is facing a great challenge.Software system itself loopholes and defects caused by abnormal behavior can make the operation and the expected behavior from each other and make the system in operation and eventually led to the collapse of the error.Therefore,it is urgent to analyze and control the abnormal behavior on the network software.The analysis theory of traditional software behavior can not be used to analyze the interactive behavior of network software.The method of combining the software and the complex network and then using the complex network perspective to study the network software is mainly focused on the network characteristics of the network software and statistical analysis of empirical research.However,the analysis and control of abnormal behavior of network software is rarely involved.Therefore,this paper deeply analyzes the propagation characteristics of abnormal behavior of network software by using the theory and method of complex network,and then puts forward a kind of control algorithm for network software system.The main research contents of this paper are as follows:(1)Based on the study of virus spreading behavior in complex networks,this paper makes a theoretical analysis and numerical simulation of abnormal behavior in the network software system by using SIR model.In this paper,we analyze and compare the infection probability,degree distribution,clustering coefficient and so on.Finally,it is concluded that these parameters have the same effect on the abnormal behavior of the network software.(2)In this paper,we creatively put forward a kind of network software local control immune algorithm based on NNVD in the network software.The algorithm starts from the anomaly source and then calculates the importance of the neighboring nodes around the abnormal source points.Then,the anomaly source is controlled by the algorithm in a certain distance around the source point,and the abnormal behavior in the network software system is finally controlled.At last,the algorithm is proved by simulation experiment,which proves that the algorithm is accurate and effective.