Automatic retraining of machine learning models to detect DDoS attacks

Patent metadata

Field	Value
Publication number	US-12244640-B2
Application number	US-202318535021-A
Country	US
Kind code	B2
Filing date	Dec 11, 2023
Priority date	Jun 29, 2016
Publication date	Mar 4, 2025
Grant date	Mar 4, 2025

Abstract

In one embodiment, a device in a network receives an attack mitigation request regarding traffic in the network. The device causes an assessment of the traffic, in response to the attack mitigation request. The device determines that an attack detector associated with the attack mitigation request incorrectly assessed the traffic, based on the assessment of the traffic. The device causes an update to an attack detection model of the attack detector, in response to determining that the attack detector incorrectly assessed the traffic.

First claim

What is claimed is:

1. A method comprising:
monitoring, at an attack detector in a network, network traffic to detect a Distributed Denial of Service (DDoS) attack by applying one or more first-level attack detection models against one or more attributes of the network traffic;
in response to detection of a DDoS attack,
causing network traffic associated with the DDoS attack to be diverted to an attack mitigation device, wherein the attack mitigation device is configured to perform a mitigation action on attack traffic in the network;
assessing, by the attack mitigation device, the network traffic associated with the DDoS attack using deep packet inspection and a second attack detection model;
providing, by the attack mitigation device, feedback to the attack detector regarding the detected DDoS attack, wherein the feedback indicates a false positive; and
refining at least one of the one or more first-level attack detection models applied by the attack detector based on the feedback.

2. The method of claim 1 wherein the feedback indicates a false negative or a false positive.

3. The method of claim 1 wherein the feedback indicates whether the attack detector incorrectly assesses the network traffic.

4. The method of claim 1 wherein refining at least one of the one or more first-level attack detection models applied by the attack detector comprises:
associating one or more labels with the traffic, based on the assessment of the traffic by the attack mitigation device; and
updating the at least one of the one or more first-level attack detection models using the one or more labels.

5. The method of claim 1 wherein refining at least one of the one or more first-level attack detection models applied by the attack detector comprises:
determining updated parameters for at least one of the one or more first-level attack detection models, based on the assessment of the traffic; and
updating the at least one of the one or more first-level attack detection models using the updated parameters.

6. The method of claim 1 , wherein the attack detector comprises a Distributed Denial of Service (DDoS) Open Threat Signaling (DOTS) client.

7. The method of claim 1 , wherein the attack mitigation device is a Distributed Denial of Service (DDoS) Open Threat Signaling (DOTS) attack mitigator.

8. The method of claim 1 wherein the attack detector comprises an attack detection device in communication with a router.

9. An apparatus, comprising:
one or more network interfaces to communicate with a network;
a processor coupled to the one or more network interfaces and configured to execute one or more processes; and
a memory configured to store a process that is executable by the processor, the process when executed operable to perform a processing comprising:
monitoring, using an attack detector, network traffic to detect a Distributed Denial of Service (DDoS) attack by applying one or more first-level attack detection models against one or more attributes of the network traffic;
in response to detection of a DDoS attack,
causing network traffic associated with the DDoS attack to be diverted to an attack mitigation device, wherein the attack mitigation device is configured to perform a mitigation action on attack traffic in the network, the mitigation action comprising: assessing the network traffic associated with the DDoS attack using deep packet inspection and a second attack detection model;
receiving feedback from an attack mitigation device regarding the detected DDoS attack, wherein the feedback indicates a false positive; and
refining at least one of the one or more first-level attack detection models applied by the attack detector based on the feedback.

10. The apparatus of claim 9 wherein the feedback indicates a false negative or a false positive.

11. The apparatus of claim 9 wherein the feedback indicates whether the attack detector incorrectly assesses the network traffic.

12. The apparatus of claim 9 wherein refining at least one of the one or more first-level attack detection models applied by the attack detector comprises:
associating one or more labels with the traffic, based on the assessment of the traffic by the attack mitigation device; and
updating the at least one of the one or more first-level attack detection models using the one or more labels.

13. The apparatus of claim 9 wherein refining at least one of the one or more first-level attack detection models applied by the attack detector comprises:
determining updated parameters for at least one of the one or more first-level attack detection models, based on the assessment of the traffic; and
updating the at least one of the one or more first-level attack detection models using the updated parameters.

14. The apparatus of claim 9 wherein the attack detector comprises a Distributed Denial of Service (DDoS) Open Threat Signaling (DOTS) client.

15. The apparatus of claim 9 wherein the apparatus is a Distributed Denial of Service (DDoS) Open Threat Signaling (DOTS) attack mitigator.

16. The apparatus of claim 9 wherein the attack detector comprises an attack detection device in communication with a router.

17. A tangible, non-transitory computer-readable medium that stores program instructions configured to cause a device in a network to execute a process comprising:
monitoring network traffic to detect a Distributed Denial of Service (DDoS) attack by applying one or more first-level attack detection models against one or more attributes of the network traffic;
in response to detection of a DDoS attack,
causing network traffic associated with the DDoS attack to be diverted to an attack mitigation device, wherein the attack mitigation device is configured to perform a mitigation action on attack traffic in the network;
assessing the network traffic associated with the DDoS attack using deep packet inspection and a second attack detection model;
providing feedback regarding the detected DDoS attack, wherein the feedback indicates a false positive; and
refining at least one of the one or more first-level attack detection models based on the feedback.

18. The tangible, non-transitory computer-readable medium of claim 17 wherein the feedback indicates a false negative or a false positive.

19. The tangible, non-transitory computer-readable medium of claim 17 wherein the feedback indicates an incorrect assessment of the network traffic.

20. The tangible, non-transitory computer-readable medium of claim 17 wherein refining at least one of the one or more first-level attack detection models comprises:
associating one or more labels with the traffic, based on the assessment of the traffic by the attack mitigation device; and
updating the at least one of the one or more first-level attack detection models using the one or more labels.

21. The tangible, non-transitory computer-readable medium of claim 17 wherein refining at least one of the one or more first-level attack detection models comprises:
determining updated parameters for at least one of the one or more first-level attack detection models, based on the assessment of the traffic; and
updating the at least one of the one or more first-level attack detection models using the updated parameters.

Assignees

• Cisco Tech Inc

Inventors

• Reddy K Tirumaleswar
• Wing Daniel G
• Anderson Blake Harrell
• Mcgrew David

Classifications

• G06N20/00

  Machine learning · CPC title

• H04L2463/144

  Detection or countermeasures against botnets · CPC title

• H04L63/1425

  Traffic logging, e.g. anomaly detection · CPC title

• H04L63/1458 Primary

  Denial of Service · CPC title