Design Of Intrusion Detection System Based On Improved CUSUM Algorithm In Cloud Environmen

The vigorous development of cloud computing technology and industrial Internet of things has provided new changes and development for the industry.Commercial cloud computing technology is the most in-depth popularization and application of the network,and industrial cloud computing technology is the necessary stage for the development of Internet of things and mobile network.Modern industrial networks benefit from the rapid development of cloud computing technology,which complement and promote each other.The industrial Internet environment produces a large amount of data information,which needs the help of cloud computing distributed management to complete the management of data information.The rapid development of cloud computing has also caused the emergence of many types of problems,especially the problem of system data security.The industrial Internet of things is in danger in terms of information security,such as the leakage of industrial core data and illegal manipulation of Internet terminals.Information security has become a major obstacle to the promotion and application of the industrial Internet of things.Therefore,it is very important to reduce the security risk of the Internet of things system and improve the security of the system.This paper summarizes the concept and service deployment mode of cloud computing,analyzes the advantages and disadvantages of cloud environment,and expounds the types of DDo S attacks that are relatively vulnerable at present.By studying the technology of intrusion detection system and combined with the environment of cloud computing,this paper designs an intrusion detection system based on improved CUSUM algorithm.This paper does the following research work mainly:(1)Firstly,this paper summarizes the definition of cloud computing,its services and deployment methods,studies and analyzes the hidden data security problems and common DDo S attacks in the cloud computing environment,and summarizes the current DDo S attack methods.This paper analyzes the common host based intrusion detection and network intrusion detection methods,deeply analyzes the concept and classification of intrusion detection system,and understands its functional modules and the functions of corresponding modules.(2)The CUSUM algorithm and the main idea of the algorithm are studied.The design idea is to accumulate the sample data information and the size displacement information of the whole process,so as to realize the amplification effect and improve the sensitivity to small displacement in the inspection process.This paper studies the cart algorithm of decision tree in machine learning,studies the basic principle of cart algorithm,and realizes data classification based on cart algorithm.(3)Based on the technology and business requirements of intrusion detection system,this paper designs a set of intrusion detection system that can be used in cloud environment,and constructs the intrusion detection system prototype through jade.The core function module of this system is based on CUSUM algorithm and cart algorithm of decision tree.The total framework,function modules and implementation platform of the system are analyzed.Simulation results show that the algorithm is efficient and the designed intrusion detection model is reliable.