Hierarchical Network Security Measurement and Optimal Proactive Defense in Cloud Computing Environments. (23rd March 2022)

• Record Type: Journal Article
• Authors: Xing, Jingyu; Zhang, Zheng
• Other Names: Arif Muhammad Academic Editor.
• Is Part Of: Security and communication networks. Volume 2022(2022)
• Publication Date: 2022-03-23
• Subjects: Computer networks -- Security measures -- Periodicals; Computer security -- Periodicals; Cryptography -- Periodicals; 005.805
• DOI: 10.1155/2022/6783223
• Languages: English
• ISSNs: 1939-0114

Abstract

Abstract : This paper presents an in-depth study and analysis of hierarchical network security measurement and optimal active defense in the cloud computing environment. All the cloud platform security-related data collected through cloud platform monitoring is collected, and then the relevant security data is summarized and analyzed, so that the specific security posture index of the cloud platform can be derived, thus providing a reference for cloud platform managers to judge the security risks of the cloud platform. It provides a reference for cloud platform managers to judge the security risks of cloud platforms. Through the cloud platform security situation awareness system, we mainly study the construction of cloud platform, the construction of security situation awareness system, and the calculation of security situation value and use, thus greatly improving the stability, security, and reliability of the cloud platform. The application of the method avoids the drawbacks of traditional network security management, which relies entirely on past data and cannot sense changes in the security state of the system in real time. At the same time, the predicted results are added to the input of the fuzzy decision-making system, improving the accuracy of the assessment. The method improves the real-time and effectiveness of network security posture prediction, increases the convergence speed and prediction accuracy of the algorithm, and avoids the occurrence of overfitting. Simulation experiments based on the internet network security posture dataset show that this research method has less prediction error than the traditional machine learning methods and other deep learning methods, has higher learning efficiency, and is more rapid, accurate, and effective in predicting the trend of network security posture in the big data environment in the future period.