Dynamic task influence estimation method for self-adaptively switching Bayes network

Abstract

The invention discloses a dynamic task influence estimation method for self-adaptively switching Bayes network. The method comprises the following steps: performing distributed tracking service calling, and automatically generating an internal dependency relation of a service domain; taking the service domain as the medium to perform upward and downward mapping, and generating the internal dependency relation of a task domain and a resource domain; according to multi-element attribute description of the task, the service and the resource, constructing an association model of the task domain-service domain-resource domain; establishing the Bayes network by using the dependency features of the association model; building an association model and Bayes network base form the time-varying features of the task domain-service domain-resource domain in the system running process, and self-adaptively switching the network model according to an actual resource scheduling policy; and training the Bayes network by utilizing priori test data, and performing dynamic task influence estimation by combining actual monitoring state data.

Description

technical field [0001] The invention relates to the technical field of information systems, in particular to a method for estimating the dynamic task impact of an adaptive switching Bayesian network. Background technique [0002] When the system suffers from external attacks or internal disturbances, how to ensure the smooth completion of ongoing tasks in the system is still a challenge. Cyber ​​attacks may seriously affect mission status, mission progress, and mission completion, and may even cause mission failure. When a system is attacked, the operational commander's greatest concern is the possibility of mission completion and the degree to which the mission is affected. The task impact estimation method is to complete this function. According to the system node status fed back by the monitoring system, the impact estimation is performed on the final task, which provides a reference for the decision-making of the combat commander and the maintenance of the system suppor...

AI Technical Summary

Problems solved by technology

The method based on the situational knowledge reference model (SKRM) can realize task impact estimation, but because it does not strictly stipulate cross-layer interconnection, SKRM lacks the ability to conduct quantitative task impact analysis

Using the Impact Dependency Graph (IDG) for impact assessment can calculate the degree of task impact to a certain extent, but it does not give a detailed method for generating dependencies in the IDG graph, and the calculation method of the logical relationship between nodes also affects the accuracy of its assessment sex

[0004] Most of the existing task impact estimation methods are qualitative evaluation methods of modeling, and methods that can achieve accurate quantitative evaluation are relatively rare, and the current methods are all for static task impact estimation.

Images