Correlating network event anomalies using active and passive external reconnaissance to identify attack information
A system and method for correlating network event anomalies to identify attack information, that identifies anomalous events within the network, identifies correlations between anomalies and other network events and resources, generates a behavior graph describing an attack pathway derived from the correlations, and determines an attack point of origin using the behavior graph.
1. A system for correlating network event anomalies to identify attack information, comprising:
a cyber-physical graph module comprising a first plurality of programming instructions stored in a memory of, and operating on a processor of, a computing device, wherein the first plurality of programming instructions, when operating on the processor, cause the computing device to create a cyber-physical graph of an organization using information about the organization, the cyber-physical graph comprising nodes representing entities associated with the organization and edges representing relationships between entities associated with the organization;
a reconnaissance engine comprising a second plurality of programming instructions stored in the memory of, and operating on the processor of, the computing device, wherein the second plurality of programming instructions, when operating on the processor, cause the computing device to:
perform a reconnaissance search using the cyber-physical graph; and
apply some or all of the results of the reconnaissance search to the cyber-physical graph to create a normal behavior model for a plurality of nodes in the cyber-physical graph; and
a directed computational graph engine comprising a third plurality of programming instructions stored in the memory of, and operating on the processor of, the computing device, wherein the third plurality of programming instructions, when operating on the processor, cause the computing device to:
receive the normal behavior model;
using the cyber-physical graph and the normal behavior model:
identify an anomalous event based on analysis of cyber-physical graph and the normal behavior model;
analyze the cyber-physical graph and the normal behavior model to identify correlations between affected nodes, the affected nodes being based on the anomalous event;
generate a behavior graph based on the identified correlations;
analyze the behavior graph to produce a dependency tree comprising causative relationships between events; and
traverse the behavior tree backward in a temporal dimension to identify a plurality of starting conditions for the anomalous event.
2. The system of claim 1 , wherein the starting conditions comprise a node identified as the point-of-origin for the anomalous event.
3. The system of claim 1 , wherein the information about the organization further comprises information about business processes within the organization.
4. The system of claim 1 , wherein the information about the organization further comprises prior loss information for the organization.
5. A method for correlating network event anomalies to identify attack information, comprising the steps of:
creating a cyber-physical graph of an organization using information about the organization, the cyber-physical graph comprising nodes representing entities associated with the organization and relationships between the entities associated with the organization;
performing a reconnaissance search using the cyber-physical graph;
applying some or all of the results of the reconnaissance search to the cyber-physical graph to create a normal behavior model for a plurality of nodes in the cyber-physical graph; and
using the cyber-physical graph and the normal behavior model:
identifying an anomalous event based on analysis of the cyber-physical graph and the normal behavior model;
analyzing the cyber-physical graph and the normal behavior model to identify correlations between affected nodes, the affected nodes being based on the anomalous event;
generating a behavior graph based on the identified correlations;
analyzing the behavior graph to produce a dependency tree comprising causative relationships between events; and
traversing the behavior tree backward in a temporal dimension to identify a plurality of starting conditions for the anomalous event.
6. The method of claim 5 , wherein the starting conditions comprise a node identified as the point-of-origin for the anomalous event.
7. The method of claim 5 wherein the information about the organization further comprises information about business processes within the organization.
8. The method of claim 5 wherein the information about the organization further comprises prior loss information for the organization.