Method and system for detecting and remediating polymorphic attacks across an enterprise

• US 10,511,616 B2
• Filed: 11/06/2018
• Issued: 12/17/2019
• Est. Priority Date: 12/09/2015
• Status: Active Grant

• Current Assignee: Check Point Software Technologies Limited
• Original Assignee: Check Point Software Technologies Limited
• Inventors: Leiderfarb, Tamara, Arzi, Lior, Pal, Anandabrata
• Primary Examiner(s): Khan, Sher A
• Application Number:
  US16/181,377
• Publication Number:
  US 20190075125A1
• Time in Patent Office: 406 Days
• CPC Class Codes: H04L 63/1416 Event detection, e.g. attack signature detection
• CPC Class Codes: H04L 63/1441 Countermeasures against malicious traffic countermeasures against attacks on cryptographic mechanisms H04L9/002
• CPC Class Codes: H04L 63/145 the attack involving the propagation of malware through the network, e.g. viruses, trojans or worms

Abstract

Disclosed are methods and systems for detecting malware and potential malware based on using generalized attack trees (generalized attack tree graphs). The generalized attack trees are based on attack trees (attack tree graphs), whose objects, such as links and vertices, have been analyzed, and some of these objects have been generalized, resulting in the generalized attack tree of the invention.

Claims

1. A method for detecting potential malware comprising:

a)

     1) obtaining an attack tree representative of an attack on a network, the attack tree formed of objects;

2) analyzing the objects to determine whether each of the objects is classified as known or unknown, in accordance with predefined criteria; and

,

3) representing the unknown objects in the attack tree as generalized objects, resulting in the creation of a first generalized attack tree from the obtained attack tree;

b) dividing the first generalized attack tree into subtrees including first generalized objects;

c) obtaining at least one subtree including second generalized objects associated with a subsequent generalized attack tree;

d) comparing at least one of the subtrees from the first generalized attack tree to the at least one subtree associated with the subsequent generalized attack tree, based on at least partial matches of the first generalized objects and the second generalized objects, the least partial matches including matching less than all of the first generalized objects with the second generalized objects; and

,

e) augmenting the first generalized attack tree by adding the second generalized objects, which do not match the first generalized objects, to the first generalized attack tree, to detect potentially unknown malware.

2. The method of claim 1, wherein the objects include links and vertices.

3. The method of claim 2, wherein the links are determined to be known.

4. The method of claim 2, wherein each of the vertices is determined to be known, unless the object type of the vertice is determined as unknown when:

a) the object type is unknown in accordance with a score;

or, b) the object type is known and malicious in accordance with a score.

5. The method of claim 4, wherein the object type includes at least one of:

a uniform resource locator (URL), a file, a process, or a registry key.

6. The method of claim 2, wherein the least partial matches includes matching all of the first generalized objects with the second generalized objects.

7. The method of claim 4, wherein the score includes at least one of:

a field score, a numeric score, and an overall score based on the sum of;

     1) field scores, and,

     2) numeric scores.

8. The method of claim 1, wherein the attack on the network occurs in at least one machine linked to the network.

9. The method of claim 1, wherein the attack on the network occurs at an endpoint of the network.