Using photonic emission to develop electromagnetic emission models

Patent metadata

Field	Value
Publication number	US-11538147-B2
Application number	US-201916684507-A
Country	US
Kind code	B2
Filing date	Nov 14, 2019
Priority date	Jul 22, 2016
Publication date	Dec 27, 2022
Grant date	Dec 27, 2022

Abstract

A method and apparatus related to developing electromagnetic emission and power models for a target device using photonic emissions thereof are provided. Data of photonic emissions of a target device during a first period of time with the target device in one or more modes is recorded. Data of electromagnetic emissions of the target device during the first period of time with the target device in the one or more modes is also recorded. The recorded data of the photonic emissions and the recorded data of the electromagnetic emissions are correlated to establish one or more electromagnetic emission models for the target device. The one or more electromagnetic emission models enable predictive analysis of emissions by the target device.

First claim

What is claimed is:

1. A method, comprising:
developing one or more electromagnetic emission models for a target device based on photonic emissions of a target device in a baseline context by:
recording data of the photonic emissions of the target device during a first period of time with the target device in one or more modes;
recording data of the electromagnetic emissions of the target device during the first period of time with the target device in the one or more modes; and
correlating (i) the recorded data of the photonic emissions having a high spatial resolution and (ii) the recorded data of the electromagnetic emissions that is location independent, to establish the one or more electromagnetic emission models for the target device.

2. The method of claim 1 , further comprising:
executing one or more programs on a target device;
measuring electromagnetic emissions of the target device during the execution of the one or more programs in a non-baseline context; and
identifying an anomaly condition associated with the one or more programs executed on the target device by comparing a result of the measuring to the one or more electromagnetic emission models developed for the target device, wherein:
the identifying of the anomaly condition associated with the one or more programs executed on the target device comprises identifying a change in execution of the one or more programs by one or more circuit blocks of an integrated circuit of the target device, activation of a malware on the target device, or both, by way of a database reference; and
the recording data of the photonic emissions of the target device comprises using Picosecond Imaging Circuit Analysis (PICA) during the first period of time with the target device in one or more modes.

3. The method of claim 2 , wherein the measuring of the electromagnetic emissions of the target device in the test context comprises recording data of the electromagnetic emissions of the target device during a second period of time after the first period of time with the target device in at least one of the one or more modes.

4. The method of claim 2 , wherein the recording of the data of the photonic emissions of the target device using PICA with the target device in the one or more modes comprises:
recording data of a first PICA emission image when the target device is in a first mode of the one or more modes;
recording data of a second PICA emission image when the target device is in a second mode of the one or more modes; and
identifying a region of interest associated with one or more circuit blocks of an integrated circuit of the target device by comparing the first PICA emission image and the second PICA emission image.

5. The method of claim 4 , wherein the correlating of the recorded data of the photonic emissions and the recorded data of the electromagnetic emissions to establish the one or more electromagnetic emission models for the target device comprises correlating the one or more circuit blocks performing one or more activities in the region of interest during the first period of time to electromagnetic emission signatures recorded during the first period of time.

6. The method of claim 1 , wherein the correlating of the recorded data of the photonic emissions and the recorded data of the electromagnetic emissions to establish the one or more electromagnetic emission models for the target device comprises:
identifying transistor switching activities in one or more circuit blocks of an integrated circuit of the target device in PICA waveforms according to the recorded data of the photonic emissions of the target device; and
correlating in a time domain the identified transistor switching activities to electromagnetic emission waveforms according to the recorded data of the electromagnetic emissions of the target device to perform one or more of:
enabling temporal aggregation of the electromagnetic emission waveforms corresponding to a targeted event,
identifying one or more time frames of interest, and
improving a signal-to-noise ratio (SNR) in electromagnetic emission waveforms associated with the target device.

7. The method of claim 1 , wherein the correlating of the recorded data of the photonic emissions and the recorded data of the electromagnetic emissions to establish the one or more electromagnetic emission models for the target device comprises:
transforming the recorded data of the photonic emissions of the target device from a time domain to a frequency domain;
identifying one or more frequency components in the recoded data of the photonic emissions of the target device in the frequency domain based on one or more changing circuit behaviors of one or more circuit blocks of an integrated circuit of the target device; and
filtering, using the identified one or more frequency components, electromagnetic emission spectra according to the recorded data of the electromagnetic emissions of the target device to perform one or more of:
isolating a frequency of interest,
improving a signal-to-noise ratio (SNR) in electromagnetic emission waveforms associated with the target device, and
developing spectral signatures associated with one or more circuit behaviors of the integrated circuit of the target device.

8. The method of claim 1 , wherein the correlating of the recorded data of the photonic emissions and the recorded data of the electromagnetic emissions to establish the one or more electromagnetic emission models for the target device comprises:
analyzing in time the data of the photonic emissions recorded during one or more circuit operations of the target device corresponding to a specific electromagnetic emission signature associated with the target device in time to identify matching photonic emission signature in time;
identifying one or more locations of the target device associated with the identified matching photonic emission signature in time; and
associating the identified one or more locations of the target device to the specific electromagnetic emission signature.

9. The method of claim 1 , wherein the correlating of the recorded data of the photonic emissions and the recorded data of the electromagnetic emissions to establish the one or more electromagnetic emission models for the target device comprises:
analyzing in frequency the data of the photonic emissions recorded during one or more circuit operations of the target device corresponding to a specific electromagnetic emission signature associated with the target device in frequency to identify matching photonic emission signature in frequency;
identifying one or more locations of the target device associated with the identified matching photonic emission signature in frequency; and
associating the identified one or more locations of the target device to the specific electromagnetic emission signature.

10. A method, comprising:
executing one or more programs on a target device;
recording data of photonic emissions of the target device using Picosecond Imaging Circuit Analysis (PICA) during a first period of time with the target device in one or more modes;
recording data of electromagnetic emissions of the target device during the first period of time with the target device in the one or more modes;
correlating the recorded data of the photonic emissions having a high spatial resolution and the recorded data of the electromagnetic emissions that is location independent to establish one or more electromagnetic emission models for the target device; and
identifying an anomaly condition associated with the one or more programs executed on the target device comprising identifying a change in execution of the one or more programs by one or more circu

Assignees

• IBM

Inventors

• Bahgat Shehata Andrea
• Song Peilin
• Stellari Franco

Classifications

• H04L63/1416 Primary

  Event detection, e.g. attack signature detection · CPC title

• H04L63/162

  at the data link layer · CPC title

• G06F30/3312

  Timing analysis · CPC title

• G01R29/0814

  Field measurements related to measuring influence on or from apparatus, components or humans (EMC, EMI and similar testing in general G01R31/001), e.g. in ESD, EMI, EMC, EMP testing, measuring radiation leakage; detecting presence of micro- or radiowave emitters; dosimetry; testing shielding; measurements related to lightning · CPC title

• H04K3/82

  related to preventing surveillance, interception or detection · CPC title