Systems and methods for inhibiting threats to a computing environment

1. A system for the detection and inhibition of insider threats to a computing environment to which the system is connected, the system comprising:

a hardware bus having instructions stored thereon that, when executed, perform the following steps:

collect incoming data from incoming signals from an external device connected to the hardware bus, the collecting of the data comprising observing a level change in a data direction in of the external device in order to capture rocket identification (PID) data, the incoming data comprising the PID data;

transform the incoming data from the incoming signals into Macrocell Interface (UTMI) or UTMI+Low Pin Interface (ULPI) data packages;

perform pre-processing on the data packages to recreate protocol data packets of the external device and to obtain timestamp information and human interface device (HID) information about the incoming data from the incoming signals, the performing of the pre-processing comprising utilizing a first finite state machine to recreate the protocol data packets from the data packages and utilizing a second finite state machine on the data packets in order to observe and validate an enumeration process of the external device;

extract feature information about the external device from the recreated protocol data packets;

classify the external device as benign or malicious based on the extracted feature information;

if the external device is classified as malicious, prevent the external device from transmitting the incoming signals to the computing environment; and

if the external device is classified as benign, allow the external device to transmit the incoming signals to the computing environment.

2. The system according to claim 1 , the feature information comprising information about at least one feature of keystroke-like actions of the external device, and

the at least one feature comprising at least one of key transition time (KTT), duration held, normalized KTT, and normalized duration held.

3. The system according to claim 1 , the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising key transition time (KTT), and

the external device being classified as malicious if the KTT of any keystroke-like action of the external device is less than a first minimum threshold time.

4. The system according to claim 3 , the first minimum threshold time being 80 milliseconds.

5. The system according to claim 1 , the instructions of the hardware bus, when executed, further creating a whitelist of known benign devices having a key transition time (KTT) of less than a first minimum threshold,

the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising KTT, and

the external device being classified as malicious if the KTT of any keystroke-like action of the external device is less than the first minimum threshold time and the external device is not on the whitelist.

6. The system according to claim 1 , the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising duration held, and

the external device being classified as malicious if the duration held of any keystroke-like action of the external device is less than a second minimum threshold time.

7. The system according to claim 1 , the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising key transition time (KTT) and duration held, and

the external device being classified as benign if the KTT of all keystroke-like actions of the external device is greater than or equal to a first minimum threshold time and the duration held of all keystroke-like actions of the external device is greater than or equal to a second minimum threshold time.

8. The system according to claim 1 , the external device being a universal serial bus (USB) device.

9. The system according to claim 8 , the external device being a USB keyboard, USB mouse, USB storage device, or USB adapter.

10. The system according to claim 1 , the classifying of the external device as benign or malicious based on the extracted feature information comprising using machine learning on the extracted feature information.

11. The system according to claim 10 , the using of machine learning comprising using a decision tree, a random forest, Naive Bayes, k-nearest neighbors, or a support vector machine on the extracted information.

12. The system according to claim 10 , the using of machine learning comprising using a decision tree on the extracted information.

13. A method for the detection and inhibition of insider threats to a computing environment to which a hardware bus is connected, the method comprising:

collecting, by the hardware bus, incoming data from incoming signals from an external device connected to the hardware bus, the collecting of the data comprising observing a level change in a data direction pin of the external device in order to capture packet identification (PID) data, the incoming data comprising the PID data;

transforming, by the hardware bus, the incoming data from the incoming signals into Macrocell Interface (UTMI) or UTMI+Low Pin Interface (ULPI) data packages;

performing, by the hardware bus, pre-processing on the data packages to recreate protocol data packets of the external device and to obtain timestamp information and human interface device (HID) information about the incoming data from the incoming signals, the performing of the pre-processing comprising utilizing a first finite state machine to recreate the protocol data packets from the data packages and utilizing a second finite state machine on the data packets in order to observe and validate an enumeration process of the external device;

extracting, by the hardware bus, feature information about the external device from the recreated protocol data packets;

classifying, by the hardware bus, the external device as benign or malicious based on the extracted feature information;

if the external device is classified as malicious, preventing, by the hardware bus, the external device from transmitting the incoming signals to the computing environment; and

if the external device is classified as benign, allowing, by the hardware bus, the external device to transmit the incoming signals to the computing environment.

14. The method according to claim 13 , the feature information comprising information about at least one feature of keystroke-like actions of the external device, and

the at least one feature comprising at least one of key transition time (KTT), duration held, normalized KTT, and normalized duration held.

15. The method according to claim 13 , further comprising creating a whitelist of known benign devices having a key transition time (KTT) of less than a first minimum threshold,

the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising KTT, and

the external device being classified as malicious if the KTT of any keystroke-like action of the external device is less than the first minimum threshold time and the external device is not on the whitelist.

16. The method according to claim 13 , the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising duration held, and

the external device being classified as malicious if the duration held of any keystroke-like action of the external device is less than a second minimum threshold time.

17. The method according to claim 13 , the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising key transition time (KTT) and duration held, and

the external device being classified as benign if the KTT of all keystroke-like actions of the external device is greater than or equal to a first minimum threshold time and the duration held of all keystroke-like actions of the external device is greater than or equal to a second minimum threshold time.

18. The method according to claim 13 , the classifying of the external device as benign or malicious based on the extracted feature information comprising using machine learning on the extracted feature information, and

the using of machine learning comprising using a decision tree, a random forest, Naive Bayes, k-nearest neighbors, or a support vector machine on the extracted information.

19. The method according to claim 13 , the external device being a universal serial bus (USB) keyboard, USB mouse, USB storage device, or USB adapter.

20. A method for the detection and inhibition of insider threats to a computing environment to which a hardware bus is connected, the method comprising:

collecting, by the hardware bus, incoming data from incoming signals from an external device connected to the hardware bus, the collecting of the data comprising observing a level change in a data direction in of the external device in order to capture packet identification PID data, the incoming data comprising the PID data;

transforming, by the hardware bus, the incoming data from the incoming signals into Macrocell Interface (UTMI) or UTMI+Low Pin Interface (ULPI) data packages;

performing, by the hardware bus, pre-processing on the data packages to recreate protocol data packets of the external device and to obtain timestamp information and human interface device (HID) information about the incoming data from the incoming signals, the performing of the pre-processing comprising utilizing a first finite state machine to recreate the protocol data packets from the data packages and utilizing a second finite state machine on the data packets in order to observe and validate an enumeration process of the external device;

extracting, by the hardware bus, feature information about the external device from the recreated protocol data packets;

creating a whitelist of known benign devices having a key transition time (KTT) of less than a first minimum threshold;

classifying, by the hardware bus, the external device as benign or malicious based on the extracted feature information;

if the external device is classified as malicious, preventing, by the hardware bus, the external device from transmitting the incoming signals to the computing environment; and

if the external device is classified as benign, allowing, by the hardware bus, the external device to transmit the incoming signals to the computing environment

the feature information comprising information about at least one feature of keystroke-like actions of the external device,

the at least one feature comprising KTT and duration held,

the external device being classified as malicious if the KTT of any keystroke-like action of the external device is less than the first minimum threshold time and the external device is not on the whitelist,

the external device being classified as malicious if the duration held of any keystroke-like action of the external device is less than a second minimum threshold time,

the external device being classified as benign if the KTT of all keystroke-like actions of the external device is greater than or equal to the first minimum threshold time and the duration held of all keystroke-like actions of the external device is greater than or equal to the second minimum threshold time,

the first minimum threshold time being 80 milliseconds,

the external device being a universal serial bus (USB) keyboard, USB mouse, USB storage device, or USB adapter,

the classifying of the external device as benign or malicious based on the extracted feature information comprising using machine learning on the extracted feature information, and

the using of machine learning comprising using a decision tree on the extracted feature information.