Skip to main content
SpringerLink
Log in

Deep Learning Classification of Motherboard Components by Leveraging EM Side-Channel Signals

  • Published:
Journal of Hardware and Systems Security Aims and scope Submit manuscript

Abstract

We show that a broad range of motherboard components can be classified based on their electromagnetic (EM) emanations using a lightweight convolutional neural network (CNN) architecture. Our random bootstrap sampling approach to cross-validation allows us to select robust models that achieve strong accuracy when classifying processor, memory, power, and ethernet integrated circuits (ICs) on internet of things (IoT) devices in two different excitation states. Our method also performs equal to or better than a k-Nearest Neighbors (k-NN) classifier baseline in all tested cases. We develop an anomaly detection procedure to flag the types of IC models not seen by the CNN in training. An analysis of signals passing through our end-to-end trained model allows us to generate insights about the discriminative features in the emanations of different components. We analyze how the convolution layers’ outputs pass through linear transformation layers to understand which of the input features are most important. Our results demonstrate that a CNN architecture can be used to classify ICs accurately using a diverse set of EM emanations while providing insights about the relevant discriminative features between components.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  1. Kae-Nune N, Pesseguier S (2013) Qualification and testing process to implement anti-counterfeiting technologies into IC packages. In 2013 Design, Automation Test in Europe Conference Exhibition (DATE), pp 1131–1136

  2. Pecht M, Tiku S (2006) Bogus: electronic manufacturing and consumers confront a rising tide of counterfeit electronics. IEEE Spectr 43(5):37–46

    Article  Google Scholar 

  3. Dogan H, Forte D, Tehranipoor MM (2014) Aging analysis for recycled fpga detection. In 2014 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFT), pp 171–176

  4. Guin U, Huang K, DiMase D, Carulli JM, Tehranipoor M, Makris Y (2014) Counterfeit integrated circuits: A rising threat in the global semiconductor supply chain. Proc IEEE 102(8):1207–1228

    Article  Google Scholar 

  5. Backes M, Dürmuth M, Gerling S, Pinkal M, Sporleder C (2010) Acoustic side-channel attacks on printers. In Proceedings of the 19th USENIX Conference on Security USA, USENIX Security 10, USENIX Association, p 20

  6. Hutter M, Schmidt JM (2013) The temperature side-channel and heating fault attacks. In: Smart Card Research and Advanced Applications: 12th International Conference (CARDIS), vol. 8419. https://doi.org/10.1007/978-3-319-08302-5_15

  7. Kocher P, Jaffe J, Jun B (1999) Differential power analysis. In Advances in Cryptology (CRYPTO), vol. 1666

  8. van Eck W (1985) Electromagnetic radiation from video display units: An eavesdropping risk? Comput Secur 4(4):269–286

    Article  Google Scholar 

  9. Callan R, Zajić A, Prvulovic M (2013) A practical methodology for measure the side-channel signal available to the attacker for instruction level events. IEEE Micro 14:1–12

    Article  Google Scholar 

  10. Agrawal D, Archambeult B (2002) The em side-channel(s). Proc. Crypto. HW and Emb. Sys. (CHES), pp 29–45

  11. Yilmaz BB, Zajić A, Prvulovic M (2019) Capacity of em side channel created by instruction executions in a processor. Processings of IEEE IEMCON, pp 1–5

  12. Alam M, Khan HA, Dey M, Sinha N, Callan R, Zajić A, Prvulovic M (2018) One&done: A single-decryption em-based attack on openssl’s constant-time blinded rsa. In Proceedings of the 27th USENIX Security Symposium

  13. Kuhn M (2004) Compromising emanations: Eavesdropping risks of computer displays. Tech Rep, Computer Laboratory, University of Cambridge

  14. Dong X, Weng H, Beetner DG, Hubing TH, Wunsch DC, Noll M, Gksu H, Moss B (2006) Detection and identification of vehicles based on their unintended electromagnetic emissions. IEEE Trans Electromagn Compat 48(4):752–759

  15. Göksu H, Wunsch D, Dong X, Kökce A, Beetner D (2018) Detection and identification of vehicles based on their spark-free unintended electromagnetic emissions. IEEE Trans Electromagn Compat 60:1594–1597. https://doi.org/10.1109/TEMC.2017.2773706

  16. Vann JM, Karnowski TP, Kerekes R, Cooke CD, Anderson AL (2018) A dimensionally aligned signal projection for classification of unintended radiated emissions. IEEE Trans Electromagn Compat 60(1):122–131

    Article  Google Scholar 

  17. Yang C, Sample AP (2016) Em-id: Tag-less identification of electrical devices via electromagnetic emissions. In 2016 IEEE International Conference on RFID (RFID), pp 1–8

  18. Laput G, Yang C, Xiao R, Sample A, Harrison C (2015) Em-sense: Touch recognition of uninstrumented, electrical and electromechanical objects. In Proceedings of the 28th Annual ACM Symposium on User Interface Software & Technology (New York, NY, USA), UIST 15, Association for Computing Machinery, pp 157-166. https://doi.org/10.1145/2807442.2807481

  19. Ahmed MM, Hely D, Perret E, Barbot N, Siragusa R, Bernier M, Garet F (2018) Authentication of microcontroller board using non-invasive em emission technique. In 2018 IEEE 3rd International Verification and Security Workshop (IVSW), pp 25–30

  20. Yilmaz BB, Mert Ugurlu E, Zajic A, Prvulovic M (2020) Cell-phone classification: A convolutional neural network approach exploiting electromagnetic emanations. In ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp 2862–2866

  21. Cobb WE (2017) Exploitation of Unintentional Information Leakage from Integrated Circuits. PhD thesis, Air Force Institute of Technology

  22. Werner FT, Yilmaz BB, Prvulovic M, Zajić A (2020) Leveraging em side-channels for recognizing components on a motherboard. IEEE Trans Electromagn Compat 1–14

  23. www.ice.gov (2011) Visiontech administrator sentenced to prison for role in sales of counterfeit circuits destined to us military. https://www.ice.gov/news/releases/visiontechadministrator-sentenced-prison-role-sales-counterfeitcircuits-destined-us. [Online; posted 25-October-2011]

  24. venturebeat.com (2011) Feds close huge chip counterfeiting case. https://venturebeat.com/2011/09/25/fedsclose-the-books-on-a-huge-chip-counterfeiting-scheme/. [Online; posted 25-September-2011]

  25. Werner F, Chu DA, Djordjevic AR, Olcan DI, Prvulovic M, Zajić A (2018) A method for efficient localization of magnetic field sources excited by execution of instructions in a processor. IEEE Trans Electromagn Compat 60(3):613–622

    Article  Google Scholar 

  26. Rubino R (2010) Wireless device identification from a phase noise prospective. Master’s thesis, University of Padova, Italy

  27. Glorot X, Bordes A, Bengio Y (2011) Deep sparse rectifier neural networks. In: Gordon G, Dunson D, Dudk M (Eds.) Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, Proceedings of Machine Learning Research 15:315–323. PMLR, Fort Lauderdale, FL, USA

  28. Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: A simple way to prevent neural networks from overfitting. J Mach Learn Res 15(56):1929–1958

    MathSciNet  MATH  Google Scholar 

  29. Olimex. A10-olinuxino-lime. Available at https://www.olimex.com/Products/OLinuXino/A10/A10-OLinuXino-LIME-n4GB/open-source-hardware (2020/05/08)

  30. Olimex. A13-olinuxino. Available at https://www.olimex.com/Products/OLinuXino/A13/A13-OLinuXino/open-source-hardware (2020/05/08)

  31. Olimex. A13-olinuxino-micro. Available at https://www.olimex.com/Products/OLinuXino/A13/A13-OLinuXino-MICRO/open-source-hardware (2020/05/08)

  32. Olimex. A20-olinuxino-lime. Available at https://www.olimex.com/Products/OLinuXino/A20/A20-OLinuXino-LIME/open-source-hardware (2020/05/08)

  33. Olimex. A20-olinuxino-lime2. Available at https://www.olimex.com/Products/OLinuXino/A20/A20-OLinuXino-LIME2/open-source-hardware (2020/05/08)

  34. Olimex. A20-olinuxino-micro. Available at https://www.olimex.com/Products/OLinuXino/A20/A20-OLinuXino-MICRO/open-source-hardware (2020/05/08)

  35. Olimex. A20-olinuxino. Available at https://www.olimex.com/Products/OLinuXino/A33/A33-OLinuXino/open-source-hardware (2020/05/08)

  36. Adibelli S, Juyal P, Nguyen LN, Prvulovic M, Zajic A (2020) Near-Field Backscattering-Based Sensing for Hardware Trojan Detection. IEEE Transactions on Antennas and Propagation 68(12):8082–8090. https://doi.org/10.1109/TAP.2020.3000562

  37. Riscure. Em probe station. Available at https://getquote.riscure.com/en/quote/2101064/em-probe-station.htm (2020/05/08)

  38. Technologies, K. M9391a pxie vector signal analyzer: 6 ghz. Available at https://www.keysight.com/en/pd-2317933-pn-M9391A/pxie-vector-signal-analyzer-1-mhz-to-3-ghz-or-6-ghz?&cc=US&lc=eng (2020/05/08)

  39. Prvulovic M, Zajić A, Callan RL, Wang CJ (2017) A method for finding frequency-modulated and amplitude-modulated electromagnetic emanations in computer systems. IEEE Trans Electromagn Compat 59(1):34–42

    Article  Google Scholar 

  40. Lee K, Lee K, Lee H, Shin J (2018) A simple unified framework for detecting out-of-distribution samples and adversarial attacks. In Adv Neural Inf Proces Syst 31:7167–7177. Curran Associates, Inc

  41. Khan HA, Sehatbakhsh N, Nguyen LN, Prvulovic M, Zajić A (2019) Malware detection in embedded systems using neural network model for electromagnetic side-channel signals. Journal of Hardware and Systems Security 3(4):305–318. https://doi.org/10.1007/s41635-01900074-w

    Article  Google Scholar 

  42. Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, Desmaison A, Kopf A, Yang E, DeVito Z, Raison M, Tejani A, Chilamkurthy S, SteinerB, Fang L, Bai J, Chintala S (2019) Pytorch: An imperative style, high-performance deep learning library. In Adv Neural Inf Proces Syst 32:8024–8035. Curran Associates, Inc

  43. Loshchilov I, Hutter F (2019) Decoupled weight decay regularization. In ICLR

  44. Cui Y, Jia M, Lin TY, Song Y, Belongie S (2019) Class-balanced loss based on effective number of samples

Download references

Acknowledgements

This work has been supported, in part, by DARPA LADS contract FA8650-16-C-7620 and ONR grant N00014-19-1-2287. The views and finding in this paper are those of the authors and do not reflect the views of DARPA or ONR.

Author information

Authors and Affiliations

  1. Georgia Institute of Technology, Atlanta, Georgia

    Erik J. Jorgensen, Frank T. Werner, Milos Prvulovic & Alenka Zajić

Authors
  1. Erik J. Jorgensen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Frank T. Werner
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Milos Prvulovic
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Alenka Zajić
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Erik J. Jorgensen.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jorgensen, E.J., Werner, F.T., Prvulovic, M. et al. Deep Learning Classification of Motherboard Components by Leveraging EM Side-Channel Signals. J Hardw Syst Secur 5, 114–126 (2021). https://doi.org/10.1007/s41635-021-00116-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s41635-021-00116-2

Keywords

Search

Navigation