Iterative Attack Graph Generation for Scalable Security Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Analyzing the security of large-scale information processing systems with densely connected computers is challenging due to the enormous number of attack paths, making it difficult to perform effective security analysis.
Innovation Solution
An analysis apparatus and method that iteratively selects entities to be analyzed, generates attack graphs, and assumes countermeasures to invalidate attack conditions, repeating the process over cycles until all entities are included, with the attack graph generation conditioned on assumed countermeasures, allowing for incremental analysis and reduction of attack paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If attack path analysis is performed on large-scale systems with densely connected computers, then comprehensive security coverage is achieved, but the number of attack paths increases enormously making analysis intractable
Solution Approach 1:
The patent divides the large-scale information processing system into multiple groups, where each group contains a subset of entities. Attack graphs are generated separately for each group rather than for the entire system at once. This segmentation reduces the complexity of individual attack graphs while maintaining comprehensive security coverage across all groups.
Solution Approach 2:
The patent introduces a hierarchical dimension to the attack graph analysis by organizing entities into groups and levels. Instead of a flat comprehensive attack graph, the system creates a multi-level structure where attack graphs at different hierarchical levels can be analyzed independently and then integrated, reducing overall complexity.
2Reliability
If all entities are included in the attack graph simultaneously, then complete security analysis is possible, but the computational burden becomes unmanageable
Solution Approach 1:
The patent segments the set of all entities into multiple groups that can be processed independently. By generating attack graphs for each group separately rather than including all entities simultaneously, the computational burden is divided into manageable portions that can be processed in parallel or sequentially with reduced time requirements.
Solution Approach 2:
The patent performs preliminary grouping and organization of entities before generating attack graphs. This preliminary action prepares the data structure to enable efficient, incremental graph generation across multiple cycles, reducing the overall analysis time compared to generating one comprehensive graph from scratch.
3Measurement precision
If the attack graph includes all possible attack paths, then thorough security assessment is achieved, but the graph becomes too large to analyze effectively
Solution Approach 1:
The patent divides the comprehensive attack graph into multiple smaller group-level attack graphs. Each graph maintains precision for its specific group while avoiding the exponential growth in size that would occur in a fully comprehensive graph. The segmented approach allows effective analysis of each portion without being overwhelmed by the total system complexity.
Solution Approach 2:
The patent generates attack graphs for individual groups rather than attempting to generate a complete attack graph for the entire system in one operation. This partial action approach produces manageable graphs that can be analyzed effectively, with the understanding that comprehensive coverage is achieved through the combination of multiple partial graphs across all groups.
Data Source
AI summary
An analysis apparatus includes an analysis target selection unit configured to select entities to be analyzed, an attack graph generation unit configured to generate an attack graph for a set of the selected entities, and a countermeasure implementation assumption unit configured to assume that a countermeasure for invalidating an attack condition is implemented in the generated attack graph. The attack graph generation unit generates the attack graph on a precondition that the countermeasure for invalidating the attack condition assumed by the countermeasure implementation assumption unit has been implemented. The analysis target selection unit selects at least one of all the entities to be analyzed and adds the selected entity in the analysis targets in each of the plurality of cycles.