Abstract
Neural networks have started proliferating in various different applications including ones where security can’t be compromised. Training of high performance neural network models involves high hardware requirement and is also very time consuming. This forces users to rely on third party companies for training the neural networks, exposing the trained model to unscrupulous hands and reducing the trustworthiness of the model. It has been reported in literature about mixing of samples of malicious Trojans with training data, the trained network being embedded with hidden functionalities, which can be triggered by specific patterns of the Trojan. Hence it is essential to understand the possibilities of Trojan attacks on local systems. This work is aimed towards proposing a Trojan model for a deep neural network (DNN) targeting FPGA platforms. Insertion of a simple Trojan in the activation module of a neuron resulted in a decrease of 26% in the efficiency of the DNN. This work brings out the need for more efficient defense mechanisms against such Trojans.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Colins D (2007) Trust in integrated circuits (tic). DARPA Solicitation BAA07-24
Chakraborty RS, Narasimhan S, Bhunia S (2009) Hardware trojan: threats and emerging solutions. In: 2009 IEEE international high level design validation and test workshop, 2009. HLDVT. IEEE, pp 166–171
Kakkara V, Balasubramanian K, Yamuna B, Mishra D, Lingasubramanian K, Murugan S (2020) A Viterbi decoder and its hardware trojan models: an FPGA-based implementation study. PeerJ Comput Sci 6:e250
Maruthi V, Balamurugan K, Mohankumar N (2020) Hardware trojan detection using power signal foot prints in frequency domain. In: 2020 international conference on communication and signal processing (ICCSP). IEEE, pp 1212–1216
Narasimhan S, Du D, Chakraborty RS, Paul S, Wolff FG, Papachristou CA, Roy K, Bhunia S (2013) Hardware trojan detection by multiple-parameter side-channel analysis. IEEE Trans Comput 62(11):2183–2195
Bhardwaj A, Di W, Wei J (2018) Deep learning essentials: your hands-on guide to the fundamentals of deep learning and neural network modeling. Packt Publishing Ltd
Yadav P, Menon N, Ravi V, Vishvanathan S, Pham TD (2022) EfficientNet convolutional neural networks-based android malware detection. Comput Secur 115:102622
Pothina H, Nagaraja K (2023) Artificial neural network and math behind it. In: Smart trends in computing and communications. Springer, pp 205–221
Nikhila S, Yamuna B, Balasubramanian K, Mishra D (2019) FPGA based implementation of a floating point multiplier and its hardware Trojan models. In: 2019 IEEE 16th India council international conference (INDICON). IEEE, pp 1–4
Amrutha J, Ajai AR (2018) Performance analysis of backpropagation algorithm of artificial neural networks in Verilog. In: 2018 3rd IEEE international conference on recent trends in electronics, information & communication technology (RTEICT). IEEE, pp 1547–1550
Vipin K (2019) Zynet: automating deep neural network implementation on low-cost reconfigurable edge computing platforms. In: 2019 international conference on field-programmable technology (ICFPT). IEEE, pp 323–326
Ribeiro M, Grolinger K, Capretz MA (2015) Mlaas: machine learning as a service. In: 2015 IEEE 14th international conference on machine learning and applications (ICMLA). IEEE, pp 896–902
Barni M, Kallas K, Tondi B (2019) A new backdoor attack in CNNS by training set corruption without label poisoning. In: 2019 IEEE international conference on image processing (ICIP). IEEE, pp 101–105
Clements J, Lao Y (2019) Hardware trojan design on neural networks. In: 2019 IEEE international symposium on circuits and systems (ISCAS). IEEE, pp 1–5
Clements J, Lao Y (2018) Backdoor attacks on neural network operations. In: 2018 IEEE global conference on signal and information processing (GlobalSIP). IEEE, pp 1154–1158
Geigel A (2013) Neural network Trojan. J Comput Secur 21(2):191–232
Gu T, Liu K, Dolan-Gavitt B, Garg S (2019) Badnets: evaluating backdooring attacks on deep neural networks. IEEE Access 7:47230–47244
Li S, Zhao BZH, Yu J, Xue M, Kaafar D, Zhu H (2019) Invisible backdoor attacks against deep neural networks. arXiv preprint arXiv:1909.02742
Haoxiang W, Smys S et al (2021) Overview of configuring adaptive activation functions for deep neural networks-a comparative study. J Ubiquitous Comput Commun Technol (UCCT) 3(01):10–22
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Vamsi Krishna, G.B., Balasubramanian, K., Yamuna, B. (2023). Hardware Trojan Modelling on a FPGA Based Deep Neural Network Accelerator. In: Shakya, S., Balas, V.E., Haoxiang, W. (eds) Proceedings of Third International Conference on Sustainable Expert Systems . Lecture Notes in Networks and Systems, vol 587. Springer, Singapore. https://doi.org/10.1007/978-981-19-7874-6_39
Download citation
DOI: https://doi.org/10.1007/978-981-19-7874-6_39
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-7873-9
Online ISBN: 978-981-19-7874-6
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)