Detection of recycled integrated circuits and system-on-chips based on degradation of power supply rejection ratio
Embodiments of the present disclosure provide methods, systems, apparatus, and computer program products are for detecting whether a suspect component such as an integrated circuit (IC) or a system-on-chip (SoC) is recycled. Specifically, various embodiments involve processing power supply rejection ratio (PSRR) data obtained from a low drop-out regulator (LDO) used for the suspect component using a recycle detection machine learning model to generate a recycle prediction. In particular embodiments, the recycle detection machine learning model is developed based at least in part on degradation of PSRRs of LDOs. Accordingly, a determination is made as to whether the suspect component is recycled based on the recycle prediction. If so, then an indication that the suspect component is recycled is provided.
1. A computer-implemented method for detecting whether a suspect component comprising at least one of an integrated circuit (IC) or a system-on-chip (SoC) is recycled, the method comprising:
processing power supply rejection ratio (PSRR) data obtained from a low dropout regulator (LDO) used for the suspect component using a recycle detection machine learning model to generate a recycle prediction, wherein the recycle detection machine learning model is developed based at least in part on degradation of PSRRs of LDOs, wherein the recycle detection machine learning model is developed using training and testing PSRR data collected from at least one of a plurality of standalone LDOs and a plurality of LDOs embedded in SoCs, wherein one or more of the at least one of the plurality of standalone LDOs and the plurality of LDOs embedded in SoCs are artificially aged;
determining whether the suspect component is recycled based at least in part on the recycle prediction; and
providing an indication that the suspect component is recycled in response to determining the component is recycled.
2. The computer-implemented method of claim 1 , wherein the training and testing PSRR data comprises training PSRR data collected from at least one labeled LDO used by a first component of a first vendor that is used to train the recycle detection machine learning model and testing PSRR data collected from at least one labeled LDO used by a second component of the first vendor that is used to test the recycle detection machine learning model.
3. The computer-implemented method of claim 1 , wherein the training and testing PSRR data comprises PSRR data collected from one or more labeled LDOs used by one or more components of a same category as the suspect component.
4. The computer-implemented method of claim 1 , wherein the training and testing PSRR data comprises training PSRR data collected from at least one labeled LDO used by a first component of a first vendor that is used to train the recycle detection machine learning model and testing PSRR data collected from at least one labeled LDO used by a second component of a second vendor that is used to test the recycle detection machine learning model.
5. The computer-implemented method of claim 1 , wherein the training and testing PSRR data comprises one or more labeled LDOs used by one or more components of a different category as the suspect component.
6. The computer-implemented method of claim 1 , wherein the training and testing PSRR data comprises PSRR data collected from a non-labeled LDO used by a component.
7. An apparatus for detecting whether a suspect component comprising at least one of an integrated circuit (IC) or a system-on-chip (SoC) is recycled, the apparatus comprising at least one processor and at least one memory including program code, the at least one memory and the program code configured to, with the processor, cause the apparatus to at least:
process power supply rejection ratio (PSRR) data obtained from a low drop-out regulator (LDO) used for the suspect component using a recycle detection machine learning model to generate a recycle prediction, wherein the recycle detection machine learning model is developed based at least in part on degradation of PSRRs of LDOs, wherein the recycle detection machine learning model is developed using training and testing PSRR data collected from at least one of a plurality of standalone LDOs and a plurality of LDOs embedded in SoCs, wherein one or more of the at least one of the plurality of standalone LDOs and the plurality of LDOs embedded in SoCs are artificially aged;
determine whether the suspect component is recycled based at least in part on the recycle prediction; and
provide an indication that the suspect component is recycled in response to determining the component is recycled.
8. The apparatus of claim 7 , wherein the training and testing PSRR data comprises training PSRR data collected from at least one labeled LDO used by a first component of a first vendor that is used to train the recycle detection machine learning model and testing PSRR data collected from at least one labeled LDO used by a second component of the first vendor that is used to test the recycle detection machine learning model.
9. The apparatus of claim 7 , wherein the training and testing PSRR data comprises PSRR data collected from one or more labeled LDOs used by one or more components of a same category as the suspect component.
10. The apparatus of claim 7 , wherein the training and testing PSRR data comprises training PSRR data collected from at least one labeled LDO used by a first component of a first vendor that is used to train the recycle detection machine learning model and testing PSRR data collected from at least one labeled LDO used by a second component of a second vendor that is used to test the recycle detection machine learning model.
11. The apparatus of claim 7 , wherein the training and testing PSRR data comprises one or more labeled LDOs used by one or more components of a different category as the suspect component.
12. The apparatus of claim 7 , wherein the training and testing PSRR data comprises PSRR data collected from a non-labeled LDO used by a component.
13. A computer program product for detecting whether a suspect component comprising at least one of an integrated circuit (IC) or a system-on-chip (SoC) is recycled, the computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions configured to:
process power supply rejection ratio (PSRR) data obtained from a low drop-out regulator (LDO) used for the suspect component using a recycle detection machine learning model to generate a recycle prediction, wherein the recycle detection machine learning model is developed based at least in part on degradation of PSRRs of LDOs, wherein the recycle detection machine learning model is developed using training and testing PSRR data collected from at least one of a plurality of standalone LDOs and a plurality of LDOs embedded in SoCs, wherein one or more of the at least one of the plurality of standalone LDOs and the plurality of LDOs embedded in SoCs are artificially aged;
determine whether the suspect component is recycled based at least in part on the recycle prediction; and
provide an indication that the suspect component is recycled in response to determining the component is recycled.
14. The computer program product of claim 13 , wherein the training and testing PSRR data comprises training PSRR data collected from at least one labeled LDO used by a first component of a first vendor that is used to train the recycle detection machine learning model and testing PSRR data collected from at least one labeled LDO used by a second component of the first vendor that is used to test the recycle detection machine learning model.
15. The computer program product of claim 13 , wherein the training and testing PSRR data comprises PSRR data collected from one or more labeled LDOs used by one or more components of a same category as the suspect component.
16. The computer program product of claim 13 , wherein the training and testing PSRR data comprises training PSRR data collected from at least one labeled LDO used by a first component of a first vendor that is used to train the recycle detection machine learning model and testing PSRR data collected from at least one labeled LDO used by a second component of a second vendor that is used to test the recycle detection machine learning model.
17. The computer program product of claim 13 , wherein the training and testing PSRR data comprises one or more labeled LDOs used by one or more components of a different category as the suspect component.
18. The computer program product of claim 13 , wherein the training and testing PSRR data comprises PSRR data collected from a non-labeled LDO used by a component.