# US 2021/0084058 A1 · Machine learning based intrusion detection system for mission critical systems

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0084058 A1
- Document ID: US20210084058A1
- Status: Published application
- Kind: A1
- Application number: 16/570,386
- Filed: 2019-09-13
- Published: 2021-03-18
- Earliest claimed: 2019-09-13
- Assignees: iS5 Communications Inc.
- Applicants: iS5 Communications Inc.
- Inventors: Tamer Soliman
- Web page: https://patentgrep.com/patent/US20210084058A1

## CPC

- H04L 63/1425: {Traffic logging, e.g. anomaly detection}
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 20/00: Machine learning
- Y04S 40/20: Information technology specific aspects, e.g. CAD, simulation, modelling, system security
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- Y04S: SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS

## Abstract

The present invention is generally directed to the detection of an intrusion event in a computer network and a control network. More particularly, the present invention provides a system, framework, architecture, etc. for intrusion detection functions for network elements and control elements used in high demand and/or mission critical environments, including, but not limited to, power stations, sub-stations, intelligent transportation systems (ITS), rail, traffic control systems, chemical, oil & gas, critical manufacturing, and industrial applications.

## Claims

### Claim 1

A system for intrusion detection for network and control elements used in a mission critical environment connected to a network, the system comprising:
(a) a traffic aggregator module for minoring and aggregating network traffic;
(b) a protocol category splitter module for receiving the mirrored and aggregated network traffic from the traffic aggregator and for splitting the mirrored and aggregated network traffic into a first and second protocol category;
(c) a first and second intrusion detection system (IDS) module for the first and second protocol categories, the first intrusion detection system module for analyzing the mirrored and aggregated network traffic from the first protocol category and for transmitting a first IDS associated data and the second intrusion detection system module for analyzing the mirrored and aggregated network traffic from the second protocol category and for transmitting a second IDS associated data; and
(d) a first and second security information and event management (SIEM) module for each of the first and second protocol categories, the first SIEM module for processing the first IDS associated data to make a first protocol category alert available to a user and the second SIEM module for processing the second IDS associated data to make a second protocol category alert available to the user.

### Claim 2 (depends on 1)

The system of claim 1 wherein the system further comprises a first and second global threat intelligence platform (GTI) for the first and second IDS modules, comprising:
(a) the first GTI for receiving the first IDS associated data and further analyzing the first IDS associated data for transmitting a first GTI response data to the first IDS module, the first IDS module utilizing the first GTI response data in the analyzing of the mirrored and aggregated network traffic of the first protocol category by the first IDS module; and
(b) the second GTI for receiving the second IDS associated data and further analyzing the second IDS associated data for transmitting a second GTI response data to the second IDS module, the second IDS module utilizing the second GTI response data in the analyzing of the mirrored and aggregated network traffic of the second protocol category by the second IDS module.

### Claim 3 (depends on 2)

The system of claim 2 wherein the first and second IDS associated data each comprises alerts data and analytics data.

### Claim 4 (depends on 3)

The system of claim 3 wherein the first and second SIEM module receive the first and second alerts data of the first and second IDS associated data.

### Claim 5 (depends on 3)

The system of claim 3 wherein the first and second GTI module receive the first and second analytics data of the first and second IDS associated data.

### Claim 6 (depends on 1)

The system of claim 1 wherein the first and second IDS module further comprises a first and second IDS agent module for analyzing the first and second mirrored and aggregated network traffic from the first and second protocol categories and for transmitting a first and second IDS agent data.

### Claim 7 (depends on 6)

The system of claim 6 wherein the system further comprises a first and second deep analytics engine (DAE) module for receiving the first and second IDS agent data and for transmitting a first and second DAE data.

### Claim 8 (depends on 7)

The system of claim 7 wherein the first and second IDS agent further comprises a first and second packet analyzer for analyzing and normalizing the mirrored and aggregated network traffic of the first and second protocol categories and for transmitting a first and second packet analyzer data.

### Claim 9 (depends on 8)

The system of claim 8 wherein the first and second packet analyzer data is further comprised f a first and second potential threat data and a first and second potential anomaly data.

### Claim 10 (depends on 9)

The system of claim 9 wherein the first and second IDS agent further comprises a first and second threat detection module for the receiving and analyzing the first and second potential threat data, and for transmitting the first and second threat detection data.

### Claim 11 (depends on 10)

The system of claim 10 wherein the first and second IDS agent further comprises a first and second anomaly detection module for receiving and analyzing the first and second potential anomaly data, and for transmitting the first and second anomaly detection data.

### Claim 12 (depends on 11)

The system of claim 11 wherein the first and second anomaly detection data is transmitted to the first and second DAE modules for further analysis to produce the first and second DAE data.

### Claim 13 (depends on 6)

The system of claim 6 wherein the traffic aggregator module, the protocol category splitter module and at least one of the first and second IDS agents exist on an edge Ethernet switch where the network and control elements connect.

### Claim 14 (depends on 7)

The system of claim 7 wherein at least one of the first and second DAE modules exists as a centralized function at a control room level.

### Claim 15 (depends on 2)

The system of claim 2 wherein at least one of the first and second GTI modules exist as a centralized function.

### Claim 16 (depends on 1)

The system of claim 1 wherein the first and second protocol categories are configurable.

### Claim 17 (depends on 16)

The system of claim 16 wherein the first and second protocol categories are configurable based on the group consisting of traffic type, device type, protocol, device, network address, and TCP/UDP port.

### Claim 18 (depends on 17)

The system of claim 17 wherein the first and second protocol categories are configurable based on OT traffic and IT traffic.

### Claim 19 (depends on 1)

The system of claim 1 wherein the traffic aggregator module exists at the network edge for allowing access to a copy of all of the network traffic.

### Claim 20 (depends on 1)

The system of claim 1 wherein at least one of the first or second IDS modules communicates at least one of the first or second protocol category alert to at least one of the first or second SIEM modules and the at least one of the first or second SIEM module displays the at least one of the first or second protocol category alert to the user and facilities the user's interaction with the at least one of the first or second protocol category alert.

### Claim 21 (depends on 1)

The system of claim 1 wherein at least one of the first and second SIEM modules having an interface to facilitate a user configuring at least one of the traffic aggregator module, the protocol category splitter module and at least one of the first and second IDS modules.

### Claim 22 (depends on 8)

The system of claim 8 wherein the first and second packet analyzer modules analyze and normalize the mirrored and aggregated network traffic and based on the results thereof directs potential threats to at least one of the first and second threat detection modules and potential anomalies to at least one of the first and second anomaly detection modules.

### Claim 23 (depends on 9)

The system of claim 9 wherein at least one of the first and second threat detection modules receives at least one of the first and second packet analyzer data for performing analytics to confirm a first threat and where the first threat has been detected, the at least one of the first and second threat detection modules issues a first threat alert.

### Claim 24 (depends on 10)

The system of claim 10 wherein at least one of the first and second anomaly detection modules receives at least one of the first and second packet analyzer data for performing analytics to confirm an anomaly and where the anomaly has been detected, the at least one of the first and second anomaly detection modules issues an anomaly alert.

### Claim 25 (depends on 7)

The system of claim 7 wherein at least one of the first and second DAE modules receives at least one of the first and second IDS associated data for performing further analytics to confirm a second threat and where the second threat has been detected, the at least one of the first and second DAE modules issues a second threat alert.

### Claim 26 (depends on 1)

The system of claim 1 wherein at least one of the first and second IDS modules employs machine learning.

### Claim 27 (depends on 10)

The system of claim 10 wherein at least one of the first and second threat detection modules utilizes a supervised learning training technique using a labelled threat database data set.

### Claim 28 (depends on 11)

The system of claim 11 wherein at least one of the first and second anomaly detection modules utilizes an unsupervised learning training technique using a clean slate data set from a lab environment and/or normal traffic operation.

### Claim 29 (depends on 7)

The system of claim 7 wherein at least one of the first and second DAE modules during the training process utilizes (i) supervised learning techniques with a labeled threat data set; and/or (ii) deep neural networks can be utilized for feature extraction.

### Claim 30 (depends on 7)

The system of claim 7 wherein at least one of the first and second DAE modules during the prediction process utilizes: (i) convolutions neural networks for deep and layered analytics; (ii) recurrent neural networks for consecutive conclusive analytics; and/or (iii) decision trees based analytics.

### Claim 31

A method for intrusion detection for network and control elements used in a mission critical environment connected to a network, the method comprising:
(a) minoring and aggregating network traffic from the network;
(b) splitting the mirrored and aggregated network traffic from step (a) into a first and second protocol category network traffic;
(c) analyzing the first protocol category network traffic from step (b) and transmitting a first IDS associated data and analyzing the second protocol category network traffic from step (b) and transmitting a second IDS associated data; and
(d) processing the first and second IDS associated data from step (c) for each of the first and second protocol categories to determine whether to make a first or second protocol category alert available to a user.

### Claim 32 (depends on 31)

The method of claim 31 wherein the method further comprises:
(i) further analyzing the first IDS associated data from step (c) to produce a first GTI response data, the first GTI response data utilized in the analyzing of the mirrored and aggregated network traffic from step (b) to determine whether to make a first protocol category alert available to a user in step (d); and
(ii) further analyzing the second IDS associated data from step (c) to produce a second GTI response data, the second GTI response data utilized in the analyzing of the mirrored and aggregated network traffic from step (b) to determine whether to make a second protocol category alert available to a user in step (d).

### Claim 33 (depends on 32)

The method of claim 32 wherein the first and second IDS associated data each comprises alerts data and analytics data.

### Claim 34 (depends on 33)

The method of claim 33 wherein the analysis of step (d) receives the first and second alerts data of the first and second IDS data.

### Claim 35 (depends on 33)

The method of claim 33 wherein the analysis of steps (i) and (ii) receives the first and second analytics data of the first and second IDS data.

### Claim 36 (depends on 31)

The method of claim 31 wherein the analysis of step (c) further comprises analyzing the first and second mirrored and aggregated network traffic from the first and second protocol categories and transmitting a first and second IDS agent data.

### Claim 37 (depends on 36)

The method of claim 36 wherein the method further comprises receiving the first and second IDS agent data and transmitting a first and second DAE data.

### Claim 38 (depends on 36)

The method of claim 36 wherein the analysis of step (c) further comprises analyzing and normalizing the mirrored and aggregated network traffic of the first and second protocol categories f step (b) and transmitting a first and second packet analyzer data where a potential threat or potential anomaly has been detected.

### Claim 39 (depends on 38)

The method of claim 38 wherein the first and second packet analyzer data is further comprised of a first and second potential threat data upon detection of the potential threat and a first and second potential anomaly data upon detection of the potential threat.

### Claim 40 (depends on 39)

The method of claim 39 wherein the analysis of step (c) further comprises receiving and analyzing the first and second potential threat data, and transmitting the first and second threat detection data.

### Claim 41 (depends on 39)

The method of claim 39 wherein the analysis of step (c) further comprises receiving and analyzing the first and second packet potential anomaly data, and transmitting the first and second anomaly detection data.

### Claim 42 (depends on 41)

The method of claim 41 wherein the first and second anomaly detection data is further analyzed to produce a first and second DAE data.

### Claim 43 (depends on 31)

The method of claim 31 wherein the first and second protocol categories are configurable based on any of the group consisting of traffic type, device type, protocol, device, network address, and TCP/UDP port.

### Claim 44 (depends on 43)

The method of claim 43 wherein the first and second protocol categories are configurable based on OT traffic and IT traffic.

### Claim 45 (depends on 31)

The method of claim 31 wherein step (d) further comprises an interface to facilitate the user configuring at least one of the steps of the process and interacting with the first and second protocol category alerts.

### Claim 46 (depends on 41)

The process of claim 41 wherein step (c) receives at least one of the first or second anomaly detection data and performs further analytics to determine whether to transmit a first or second protocol category alert.

### Claim 47 (depends on 46)

The method of claim 46 wherein if the further analytics cannot determine whether to transmit a first or second protocol category alert, then step (c) further comprises analyzing at least one of the first or second DAE data to produce a GTI response data, the GTI response data utilized in further analyzing the least one of the first or second anomaly detection data to determine whether to transmit a first or second protocol category alert.

### Claim 48 (depends on 31)

The method of claim 31 wherein step (c) employs machine learning.

### Claim 49 (depends on 40)

The method of claim 40 wherein step (c) employs machine learning and utilizes a supervised learning training technique using a labelled threat database data set.

### Claim 50 (depends on 41)

The method of claim 41 wherein step (c) employs machine learning and utilizes an unsupervised learning training technique using a clean slate data set from a lab environment and/or normal traffic operation.

### Claim 51 (depends on 42)

The method of claim 42 wherein step (c) employs machine learning and during the training process utilizes (i) supervised learning techniques with a labeled threat data set; and/or (ii) deep neural networks can be utilized for feature extraction.

### Claim 52 (depends on 42)

The method system of claim 42 wherein step (c) employs machine learning and during the prediction process utilizes: (i) convolutions neural networks for deep and layered analytics; (ii) recurrent neural networks for consecutive conclusive analytics; and/or (iii) decision trees based analytics.

### Claim 53

A device for intrusion detection for network and control elements used in a mission critical environment connected to a network, the device comprising:
(a) an Ethernet switch;
(b) a traffic aggregator module connected to the Ethernet switch for mirroring and aggregating network traffic;
(c) a protocol category splitter module connected to the traffic aggregator for receiving the mirrored and aggregated network traffic from the traffic aggregator and splitting the mirrored and aggregated network traffic into a first and second protocol category; and
(d) a first and second intrusion detection system (IDS) module for the first and second protocol categories, the first intrusion detection system module for analyzing the mirrored and aggregated network traffic from the first protocol category and for transmitting a first IDS associated data and the second intrusion detection system module for analyzing the mirrored and aggregated network traffic from the second protocol category and for transmitting a second IDS associated data; wherein the first and second IDS associated data is processed to determine whether to make a first or second protocol category alert available to a user.

### Claim 54 (depends on 53)

The device of claim 53 further comprising wherein the first and second IDS associated data each comprises alerts data and analytics data.

### Claim 55 (depends on 54)

The device of claim 54 wherein the first and second IDS module further comprises a first and second IDS agent module for analyzing the first and second mirrored and aggregated network traffic from the first and second protocol categories and for transmitting a first and second IDS agent data.

### Claim 56 (depends on 55)

The device of claim 55 wherein the first and second IDS agent further comprises a first and second packet analyzer for analyzing and normalizing the mirrored and aggregated network traffic of the first and second protocol categories and for transmitting first and second packet analyzer data.

### Claim 57 (depends on 56)

The device of claim 56 wherein the first and second packet analyzer data is further comprised of a first and second potential threat data and a first and second potential anomaly data.

### Claim 58 (depends on 57)

The device of claim 57 wherein the first and second IDS agent further comprises a first and second threat detection module for the receiving and analyzing the first and second potential threat data, and for transmitting the first and second threat detection data.

### Claim 59 (depends on 57)

The device of claim 57 wherein the first and second IDS agent further comprises a first and second anomaly detection module for receiving and analyzing the first and second potential anomaly data, and for transmitting the first and second anomaly detection data.

### Claim 60 (depends on 59)

The device of claim 59 wherein the device further comprises a first and second deep analytics engine (DAE) module for receiving the first and second IDS anomaly detection data and for further analyzing the first and second IDS anomaly detection data and transmitting a first and second DAE data.

### Claim 61 (depends on 53)

The device of claim 53 wherein the first and second protocol categories are configurable based on the group consisting of traffic type, device type, protocol, device, network address, and TCP/UDP port.

### Claim 62 (depends on 61)

The device of claim 61 wherein the first and second protocol categories are configurable based on OT traffic and IT traffic.

### Claim 63 (depends on 59)

The device of claim 59 wherein the first and second packet analyzer modules for analyzing and normalizing the mirrored and aggregated network traffic and based on the results of the analysis directs potential threats to at least one of the first and second threat detection modules and potential anomalies to at least one of the first and second anomaly detection modules.

### Claim 64 (depends on 58)

The device of claim 58 wherein at least one of the first and second threat detection modules receives at least one of the first and second potential threat data for performing analytics to confirm a first threat and where the first threat has been detected, the at least one of the first and second threat detection modules issues a first threat alert.

### Claim 65 (depends on 59)

The device of claim 59 wherein at least one of the first and second anomaly detection modules receives at least one of the first and second potential anomaly data for performing analytics to confirm an anomaly and where the anomaly has been detected, the at least one of the first and second anomaly detection modules issues an anomaly alert.

### Claim 66 (depends on 60)

The device of claim 60 wherein at least one of the first and second DAE modules receives at least one of the first and second IDS agent associated data for performing further analytics to confirm a threat and where the threat has been detected, the at least one of the first and second DAE modules issues a threat alert.

### Claim 67 (depends on 53)

The device of claim 53 wherein at least one of the first and second IDS modules employs machine learning.

### Claim 68 (depends on 58)

The device of claim 58 wherein at least one of the first and second threat detection modules utilizes a supervised learning training technique using a labelled threat database data set.

### Claim 69 (depends on 59)

The device of claim 59 wherein at least one of the first and second anomaly detection modules utilizes an unsupervised learning training technique using a clean slate data set from a lab environment and/or normal traffic operation.

### Claim 70 (depends on 60)

The device of claim 60 wherein at least one of the first and second DAE modules during the training process utilizes (i) supervised learning techniques with a labeled threat data set; and/or (ii) deep neural networks can be utilized for feature extraction.

### Claim 71 (depends on 60)

The device of claim 60 wherein at least one of the first and second DAE modules during the prediction process utilizes: (i) convolutions neural networks for deep and layered analytics; (ii) recurrent neural networks for consecutive conclusive analytics; and/or (iii) decision trees based analytics.

### Claim 72 (depends on 53)

The device of claim 53 wherein the Ethernet switch further comprises an Ethernet line card connected to a switch fabric.

### Claim 73 (depends on 72)

The device of claim 72 wherein the switch fabric further comprises a switch backplane.

### Claim 74 (depends on 73)

The device of claim 73 wherein the traffic aggregator is connected to the switch backplane, is configurable to selectively mirrors traffic of at least one port in both the ingress and egress directions, and has no impact on traffic between end devices.
