High-fidelity model-driven deception platform for cyber-physical systems

1. A method for protecting a cyber-physical system against a potential attacker, the method comprising:

collecting historical information about the cyber-physical system;

training, based on the historical information, a machine-learned model to predict future conditions of at least a portion of the cyber-physical system;

generating a decoy component configured to simulate, based on the future conditions predicted by the machine-learned model, a functionality of the at least a portion of the cyber-physical system;

configuring the decoy component to:

detect an input signal indicative of an attempt by the potential attacker to gain access to or control over the cyber-physical system; and

responsive to detecting the input signal, perform operations of:

output an alert to the cyber-physical system indicative of a possible attack at the decoy component; and

respond to the input signal by simulating the functionality of the at least a portion of the cyber-physical system to distract the potential attacker; and

deploying the decoy component within the cyber-physical system.

2. The method of claim 1 , further comprising:

integrating the decoy component with other components of the cyber-physical system,

wherein the decoy component simulates the functionality of the at least a portion of the cyber-physical system by exchanging information with at least one of the other components of the cyber-physical system.

3. The method of claim 2 , wherein:

the information exchanged with the at least one of the other components of the cyber-physical system includes a variable value based on the future conditions predicted by the machine-learned model.

4. The method of claim 1 , wherein: the decoy component simulates the functionality of the at least a portion of the cyber-physical system by maintaining variables that dynamically change based on the future conditions predicted by the machine-learned model.

5. The method of claim 4 , further comprising:

responding, via the decoy component, to the input signal by outputting information indicative of one or more of the variables that dynamically change based on the future conditions predicted by the machine-learned model.

6. The method of claim 1 , further comprising:

deploying the decoy component dynamically to intercept the input signal.

7. A system for protecting a cyber-physical system against a potential attacker of the cyber-physical system, comprising at least one processor configured to:

collect historical information about the cyber-physical system;

train, based on the historical information, a machine-learned model to predict future conditions of at least a portion of the cyber-physical system; and

generate a decoy component configured to simulate, based on the future conditions predicted by the machine-learned model, a functionality of the at least a portion of the cyber-physical system;

configure the decoy component to:

detect an input signal indicative of an attempt by the potential attacker to gain access to or control over the cyber-physical system; and

responsive to detecting the input signal, perform operations of:

output an alert to the cyber-physical system indicative of a possible attack at the decoy component; and

respond to the input signal by simulating the functionality of the at least a portion of the cyber-physical system to distract the potential attacker; and

deploy the decoy component within the cyber-physical system.

8. The system of claim 7 , wherein the at least one processor is configured to simulate the functionality and communications of the at least a portion of the cyber-physical system by generating network traffic associated with the cyber-physical system.

9. The system of claim 8 , wherein the at least one processor is configured to generate the network traffic associated with the cyber-physical system by outputting decoy communications between the decoy component of the cyber-physical system and at least one other component of the cyber-physical system.

10. The system of claim 7 , wherein the machine-learned model comprises a neural network.

11. The system of claim 7 , wherein the at least one processor is further configured to:

integrate the decoy component into the cyber-physical system that simulates, based on the future conditions predicted by the machine-learned model, the functionality and communications of the at least a portion of the cyber-physical system.

12. The system of claim 11 , wherein the at least one processor is configured to integrate the decoy component into the cyber-physical system by registering the decoy component.

13. The system of claim 12 , wherein the at least one processor is configured to output the alert to the cyber-physical system further in response to determining that the decoy component is registered as a decoy of cyber-physical system.

14. The system of claim 7 , wherein the at least one processor is configured to simulate the functionality and communications of the at least a portion of the cyber-physical system by outputting information indicative of one or more of variables that dynamically change based on the future conditions predicted by the machine-learned model.

15. The system of claim 7 , wherein the machine-learned model is continuously trained, based on future information collected about the cyber-physical system, to predict the future conditions of the at least a portion of the cyber-physical system.

16. A non-transitory computer-readable media comprising instructions that, when executed, configure a processor to:

collect historical information about a cyber-physical system;

train, based on the historical information, a machine-learned model to predict future conditions of at least a portion of the cyber-physical system; and

generate a decoy component configured to simulate, based on the future conditions predicted by the machine-learned model, a functionality of the at least a portion of the cyber-physical system;

configure the decoy component to:

detect an input signal indicative of an attempt by a potential attacker to gain access to or control over the cyber-physical system; and

responsive to detecting the input signal, perform operations of:

output an alert to the cyber-physical system indicative of a possible attack at the decoy component; and

respond to the input signal by simulating the functionality of the at least a portion of the cyber-physical system to distract the potential attacker; and

deploy the decoy component within the cyber-physical system.

17. The computer-readable media of claim 16 , wherein the instructions, when executed, further configure the processor to simulate the functionality and communications of the at least a portion of the cyber-physical system by generating network traffic associated with the cyber-physical system.

18. The computer-readable media of claim 17 , wherein the instructions, when executed, further configure the processor to generate the network traffic associated with the cyber-physical system by outputting decoy communications between the decoy component of the cyber-physical system and at least one other component of the cyber-physical system.

19. The computer-readable media of claim 16 , wherein the machine-learned model comprises a neural network.

20. The computer-readable media of claim 16 , wherein the instructions, when executed, further configure the processor to integrate the decoy component into the cyber-physical system that simulates, based on the future conditions predicted by the machine-learned model, the functionality and communications of the at least a portion of the cyber-physical system.