ELECTRONIC COMPONENT CLASSIFICATION

• US 20160282394A1
• Filed: 06/10/2016
• Published: 09/29/2016
• Est. Priority Date: 12/13/2013
• Status: Active Grant

• Current Assignee: Battelle Memorial Institute
• Original Assignee: Battelle Memorial Institute
• Inventors: Engelhart, Dale C., House, Larry J.
• Granted Patent: US 9,759,757 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01R 29/26 Measuring noise figure; Measuring signal-to-noise ratio
• CPC Class Codes: G06F 18/24147 Distances to closest patterns, e.g. nearest neighbour classification
• CPC Class Codes: G06F 21/44 Program or device authentication
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G06V 20/80 Recognising image objects characterised by unique random patterns

Abstract

A system and method of electronic component authentication or component classification can reduce the vulnerability of systems (e.g., satellites, weapons, critical infrastructure, aerospace, automotive, medical systems) to counterfeits. Intrinsic deterministically random property data can be obtained from a set of authentic electronic components, processed, and clustered to create a classifier that can distinguish whether an unknown electronic component is authentic or counterfeit.

Claims

1. A method of detecting a counterfeit electronic component comprising:

temporarily installing a plurality of known electronic components from one or more trusted sources into a component interface of an electronic component authentication system for signal measurement;

measuring, during an idle state, an intrinsic deterministically random noise signal for each of the plurality of known electronic components via the component interface;

building a classifier based on the intrinsic deterministically random noise signals of the plurality of known electronic components;

non-destructively classifying an unknown electronic component from an untrusted source using the classifier.

2. The method of detecting a counterfeit electronic component of claim 1 wherein building a classifier includes:

segmenting each intrinsic deterministically random noise signal into an intrinsic deterministically random noise vector;

transforming each intrinsic deterministically random noise vector into a feature vector;

conducting a statistical analysis on each feature vector; and

clustering the feature vectors to create the classifier.

3. The method of detecting a counterfeit electronic component of claim 2 wherein the feature vector is time independent and spatially independent.

4. The method of detecting a counterfeit electronic component of claim 2 wherein the feature vector is in the frequency domain.

5. The method of detecting a counterfeit electronic component of claim 2 wherein said conducting the statistical analysis on the feature vector includes organizing the feature vector by variance and discarding dimensions where the variance is below a threshold.

6. The method of detecting a counterfeit electronic component of claim 2 wherein said conducting the statistical analysis on the feature vector includes organizing the feature vector by variance and discarding all but a predefined number of dimensions that have the highest variance.

7. The method of detecting a counterfeit electronic component of claim 2 wherein said clustering includes selecting a clustering algorithm based on properties of said unknown electronic components.

8. The method of detecting a counterfeit electronic component of claim 2 wherein said clustering includes comparing the feature vectors to each other using a distance metric.

9. The method of detecting a counterfeit electronic component of claim 8 wherein the distance metric includes Euclidean distance.

10. The method of detecting a counterfeit electronic component of claim 8 wherein the distance metric includes cosine distance.

11. The method of detecting a counterfeit electronic component of claim 8 wherein transforming each intrinsic deterministically random noise vector into a feature vector using a hybrid transformation, wherein said hybrid transformation includes transforming each intrinsic deterministically random noise vector using a plurality of different transformations that are combined to create each feature vector.

12. The method of detecting a counterfeit electronic component of claim 1 wherein each of the plurality of known electronic components belongs to one of a plurality of chip classes.

13. The method of detecting a counterfeit electronic component of claim 1 wherein measuring an intrinsic deterministically random noise signal for each of a plurality of known electronic components from one or more trusted sources includes measuring at least one of a current signal and a power signal.

26. The method of detecting a counterfeit electronic component of claim 1 wherein the component interface includes connections for rail voltage, ground, and an external clock.

14. A system for classifying an unknown electronic component comprising:

an electronic component interface for electrically interfacing with the unknown electronic component;

a sensor for measuring a power consumption signal of the unknown electronic component during an idle state via said electronic component interface;

a control unit configured to;

process a power consumption signal of the unknown electronic component to generate a noise signature of the unknown electronic component;

retrieve previously-stored noise signature information extracted from power consumption signals measured during an idle state of trusted authentic electronic components;

compare said noise signature of the unknown electronic component and said previously-stored noise signature information; and

classify the unknown electronic component based on the comparison.

15. The system of claim 14 wherein said unknown electronic component is at least one of a passive component, an analog component, and a digital component.

16. The system of claim 14 wherein the system includes an apparatus for generating said previously-stored noise signature information, said apparatus configured to:

measure an intrinsic deterministically random noise signal for each of a plurality of electronic components during an idle state, wherein the plurality of electronic components are authenticated or obtained from a trusted source;

segment each intrinsic deterministically random noise signal into an intrinsic deterministically random noise vector;

transform each intrinsic deterministically random noise vector into a feature vector;

conduct a statistical analysis on each feature vector; and

cluster the feature vectors to generate noise signature information.

17. The system of claim 16 wherein said feature vector is time independent and spatially independent.

18. The system of claim 16 wherein said feature vector is in the frequency domain.

19. The system of claim 16 wherein said statistical analysis on the feature vector includes organizing the feature vector by variance and discarding dimensions where the variance is below a threshold.

20. The system of claim 16 wherein said statistical analysis on the feature vector to includes organizing the feature vector by variance and discarding all but a predefined number of dimensions that have the highest variance.

21. The system of claim 16 wherein said clustering the feature vectors includes shared nearest neighbor clustering.

22. The system of claim 16 wherein said clustering includes comparing the feature vectors to each other using a distance metric.

23. The system of claim 22 wherein the distance metric includes cosine distance.

24. The system of claim 16 wherein said apparatus is configured to transform each intrinsic deterministically random noise vector into a feature vector using a hybrid transformation, wherein said hybrid transformation includes transforming each intrinsic deterministically random noise vector using a plurality of different transformations that are combined to create each feature vector.

25. The system of claim 16 wherein measuring an intrinsic deterministically random noise signal for each of a plurality of known electronic components from one or more trusted sources includes measuring at least one of a current signal and a power signal.

27. The system of claim 14 wherein the component interface includes connections for rail voltage, ground, and an external clock.

28. A method of classifying a plurality of electronic components comprising:

temporarily installing a plurality of electronic components into a component interface of an electronic component authentication system for signal measurement;

measuring, during an idle state, a power consumption signal for each electronic component via the component interface;

processing the power consumption signal of each electronic component to generate a deterministically random noise signature of each of the plurality of electronic components;

classifying the plurality of electronic components into different classes based on similarity of the deterministically random noise signatures.

29. The method of claim 28 including:

segmenting each intrinsic deterministically random noise signature into an intrinsic deterministically random noise vector;

transforming each intrinsic deterministically random noise vector into a feature vector;

conducting a statistical analysis on each feature vector; and

clustering the feature vectors to create the different classes.

30. The method of claim 28 wherein at least some of the different classes represent different generations of electronic components.