Font Size: a A A

Research On Endogenous Security Methods Of Host System

Posted on:2022-09-15Degree:MasterType:Thesis
Country:ChinaCandidate:Z W JiangFull Text:PDF
GTID:2518306740494574Subject:Cyberspace security
Abstract/Summary:PDF Full Text Request
At present,attacks on various information systems for cyberspace security are becoming more diverse and intense.In addition,the security of important information systems is threatened.The original security system is expert in preventing risks,but the shortcomings of lack of initiative and flexibility are gradually exposed.The high adaptability and strong evolution of biological systems can be applied to fill this gap in capabilities and provide new ideas for cyberspace security.This paper proposes a endogenous security system based on the bionic immune mechanism,studies the human immune and neural reflex mechanisms,and arranges neurons in each part of the information system,so that every basic module in the system has both functional and security characteristics.In the process of task execution,the information of the modules is continuously collected through neurons,and the resources of the information system are uniformly allocated or partially optimized through the brain analysis and processing.The cooperation between the modules is used to jointly resist risks and ensure the execution of tasks.The decision-making layer of the system sets strategies to solve and even avoid problems.Through the analysis of the information system model,it shows that this endogenous security system based on bionic immunity is effective,and the correctness of the theory is supported by constructing a prototype system and analyzing the operating conditions of the system under different strategies and the comparison results of the response time and stability of the endogenous security system and traditional protection technology when they are attacked.This thesis chooses the commonly used information system to model it,imitates the human neural control system and establishes a task-oriented endogenous safety control architecture.The correctness of the model was proved through formal language,and corresponding anomaly detection and error correction algorithms were proposed.At the same time,a prototype system based on the endogenous safety method was designed and implemented,which verified the effectiveness of the method proposed in this paper.The main work of this article is as follows:1.Study the details of the human immune mechanism and neural control mechanism,analyze the similarities and differences between the human body system and the information system,model the existing information system,and propose a task-oriented endogenous security control architecture.2.Aiming at the above model,a formal definition of the endogenous safety system model is made by decomposing the task into a series of basic action sequences,and the safe operability of the endogenous safety model is proved by the proof method based on the non-interference model..3.Summarize the abnormal behavior of the endogenous security system,make a mathematical definition of the abnormal behavior of the system,and propose an abnormal behavior detection algorithm and an abnormal behavior error correction algorithm.The experimental results show how the information system adjusts the overall operating strategy and the coordination of resources collaboration in response to exceptions.It has better performance when the types of safety parameters are not single.4.Designed and implemented a prototype system based on endogenous security methods,including communication modules,encryption and decryption modules,and interaction modules.The system operation information is collected through a large number of neurons arranged in the modules,and the system resources are uniformly deployed after analysis and processing by the control center to respond to attacks and complete tasks.Finally,the test verifies the correctness of the system model and the effectiveness of the algorithm.
Keywords/Search Tags:Endogenous safety, Bionic immunity, Function and Safety, Model Analysis and Verification, Resource collaboration
PDF Full Text Request
Related items