Network Security Real-Time Risk Assessment Model And Algorithm

In the information security engineering, Risk Assessment plays an important part. It is an essential element of Risk Management. The purpose of the research in this article is to explore the model of Real-Time Risk Assessment and to improve the frequency and precision of Risk Assessment. We are struggling to find efficient ways.Information Security Risk Assessment is a means of providing decision-makers with information needed to understand the vulnerabilities and threats factors that can negatively influence operations and outcomes and make informed judgments concerning the extent of actions needed to reduce risk.Assume that the security of the asset which consist of network and host and the useful information can be modeled by N states. There is a multitude of sensors that can provide security relevant information such as IDS. The security states of assets should be dynamically updated based on these security relevant observable information. Risks is regarded as performance function of state.Traditional methods are most manual, but it become very difficult to handle the high-speed traffic. In this article we presents a quantitative approach for state estimation. The application for HMMs and DBNs meet the requirement.In this article, there are two outstanding contributions. The first is to improve the parameter learning algorithm of CTHMMs. We compare it to the previous using the following indicators: pattern recognition and error rate as well as algorithm complexity. Second, we firstly use the DBNs to carry out risk assessments. The error rate and the algorithm complexity are greatly reduced. This is also the innovation of this article.Improved algorithm is a contribution to the Markov theory. Stochastic statistical model pertain to Discrete event dynamic system has recently become a familiar tool in computer and network security research. The importance and significance is difficult to express with ordinary glossary.