System for identifying illegitimate communications between computers by comparing evolution of data flows

• US 9,847,924 B2
• Filed: 04/02/2015
• Issued: 12/19/2017
• Est. Priority Date: 10/10/2012
• Status: Active Grant

• Current Assignee: Lancaster University Business Enterprises Limited
• Original Assignee: Lancaster University Business Enterprises Limited
• Inventors: Angelov, Plamen, Bruncak, Radovan, Hutchison, David, Simpson, Steven, Smith, Paul
• Primary Examiner(s): Luu, Le H
• Application Number:
  US14/677,283
• Publication Number:
  US 20150304198A1
• Time in Patent Office: 992 Days
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level
• CPC Class Codes: H04L 43/12 Network monitoring probes
• CPC Class Codes: H04L 63/1408 by monitoring network traffic monitoring network traffic per se H04L43/00
• CPC Class Codes: H04L 63/1425 Traffic logging, e.g. anomaly 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
• CPC Class Codes: H04L 63/16 Implementing security features at a particular protocol layer

Abstract

A method and apparatus for identifying similar and coordinated communications between computers connected by a network are described. Communications between a plurality of pairs of computers are monitored to obtain respective flow metrics for a first and second pair of computers. The flow metric represents at least one property of the data flow between the pair of computers. Representations of the evolution of the data flows between the pairs of computers are updated using the flow metrics. The representations of the evolution of the data flows are compared to determine the similarity of the data flows between the pairs of computers. The first pair of computers and the second pair of computers are identified as exhibiting similar and coordinated communication if their data flows are determined to be similar.

Claims

1. A real-time method of identifying similar and coordinated communications between a plurality of computers connected by a network, the method comprising:

monitoring communications between a plurality of pairs of computers over the network to obtain a first flow metric for a first pair of computers and a second flow metric for a second pair of computers, wherein the first flow metric represents at least one property of a first data flow between the first pair of computers and the second flow metric represents at least one property of a second data flow between the second pair of computers;

updating a representation of the evolution of the first data flow between the first pair of computers using the first flow metric or updating a representation of the evolution of the second data flow between the second pair of computers using the second flow metric;

comparing the representation of the evolution of the first data flow and the representation of the evolution of the second data flow to determine the similarity of the first data flow and the second data flow; and

identifying the first pair of computers and the second pair of computers as exhibiting similar and coordinated communication if the first data flow and second data flow are determined to be similar.

2. The method of claim 1, wherein the flow metric includes the average number of bytes per unit time transmitted between the pair of computers.

3. The method of claim 1, wherein the flow metric includes the average number of bytes per packet transmitted between the pair of computers.

4. The method of claim 1, and further comprising:

identifying computers that are a source of similar and coordinated communication; and

only monitoring communications between a pair of computers if at least one of the computers of the pair has been identified as a source of similar and coordinated communication.

5. The method of claim 1, and further comprising:

using a self-organizing map to arrange the representations of the evolution of the data flow by similarity, and wherein only representations within a limited range of similarity are compared.

6. The method of claim 1, wherein clustering is used and wherein the representation of the evolution of the data flow comprises a sequence of clusters of the flow metric.

7. The method of claim 6, wherein comparing the representations includes determining if any cluster of the representation of the evolution of the first data flow matches more than one cluster of the representation of the evolution of the second data flow.

8. The method of claim 7, wherein a match between clusters is determined based on the separation of the centres of the clusters, the support of the clusters or the radius of the clusters.

9. The method of claim 8, wherein remediating communication comprises one or more of:

blocking;

filtering, and switching.

10. The method of claim 6, wherein comparing the representations includes determining if a pair of first clusters and a pair of second clusters of the representation of the evolution of the first data flow and the representation of the evolution of the second data flow match.

11. The method of claim 1, further comprising:

remediating communication between the first pair of computers and/or the second pair of computers if they are identified as exhibiting illegitimate communication.

12. The method of claim 1, further comprising clearing a representation of the evolution of a data flow from memory when the data flow is determined unlikely to correspond to illegitimate communication.

13. The method of claim 12, wherein the representation of the evolution is removed from memory if the age of the representation of the evolution is greater than the age of the representation of the evolution of the first data flow or second data flow for the first pair of computers and the second pair of computers that have been identified as exhibiting illegitimate communication.

14. The method of claim 12, wherein the representation of the evolution is removed from memory a fixed period of time after the first pair of computers and the second pair of computers that have been identified as exhibiting illegitimate communication.

15. The method of claim 1, wherein each data flow over the network is treated separately at a network level by a tuple including a source IP address and a destination IP address.

16. The method of claim 1, wherein the method identifies botnets and pairs of computers exhibiting botnet communication behaviour.

17. A data processing apparatus comprising one or more data processing devices and one or more non-transitory computer readable media, the non-transitory computer readable medium or media storing computer program code executable by the data processing device or devices to carry out a real-time method of identifying similar and coordinated communications between a plurality of computers connected by a network, the method comprising:

monitoring communications between a plurality of pairs of computers over the network to obtain a first flow metric for a first pair of computers and a second flow metric for a second pair of computers, wherein the first flow metric represents at least one property of a first data flow between the first pair of computers and the second flow metric represents at least one property of a second data flow between the second pair of computers;

updating a representation of the evolution of the first data flow between the first pair of computers using the first flow metric or updating a representation of the evolution of the second data flow between the second pair of computers using the second flow metric;

comparing the representation of the evolution of the first data flow and the representation of the evolution of the second data flow to determine the similarity of the first data flow and the second data flow; and

identifying the first pair of computers and the second pair of computers as exhibiting similar and coordinated communication if the first data flow and second data flow are determined to be similar.

18. The data processing apparatus of claim 17, wherein the data processing apparatus includes or comprises a router.

19. The data processing apparatus of claim 17, wherein the apparatus is distributed over different physical devices which are in communication.

20. One or more non-transitory computer readable media storing computer program code executable by one or more data processing devices to carry out a real-time method of identifying similar and coordinated communications between a plurality of computers connected by a network, the method comprising:

monitoring communications between a plurality of pairs of computers over the network to obtain a first flow metric for a first pair of computers and a second flow metric for a second pair of computers, wherein the first flow metric represents at least one property of a first data flow between the first pair of computers and the second flow metric represents at least one property of a second data flow between the second pair of computers;

updating a representation of the evolution of the first data flow between the first pair of computers using the first flow metric or updating a representation of the evolution of the second data flow between the second pair of computers using the second flow metric;

comparing the representation of the evolution of the first data flow and the representation of the evolution of the second data flow to determine the similarity of the first data flow and the second data flow; and

identifying the first pair of computers and the second pair of computers as exhibiting similar and coordinated communication if the first data flow and second data flow are determined to be similar.