Information Leakage Mitigation to Protect the Convolutional Neural Networks Against the Remote Side-Channel Analysis
Articles in Press, Accepted Manuscript, Available Online from 15 May 2026
https://doi.org/10.22042/isecure.2026.243620
Farid Rajabzadeh, Ali Jahanian
Abstract Machine learning systems, despite exhibiting high inference accuracy in practical applications, are susceptible to security and reliability concerns both during the training phase and the inference phase. In this paper, we have demonstrated that it is possible to extract internal information from a neural network without physical access. This attack was executed through the utilization of a power sensor. This sensor enables remote sampling. Thus far, the sensor has been employed to extract power samples from cryptographic circuits, and its functionality and correctness have been thoroughly tested. Now, in this paper, the same power sensor is used to extract power samples from a neural network, allowing us to assess the supervisor’s performance for applications beyond cryptographic algorithms. In this paper, we demonstrate that the power sensor accurately extracts power samples from neural networks. This paper reveals that between 20,000 and 50,000 power samples of a 16-bit neural network weight can be retrieved. The final step involved hardening the neural network against side-channel attacks. Test results in this section demonstrate that it is possible to make the neural network resistant to first-order side-channel attacks with an area overhead of about 6%. The degree of reinforcement was measured using the assumption test method, revealing that the attack has become eight times more challenging.
Side channel parameter characteristics of code injection attacks
Volume 9, Issue 1, January 2017, Pages 17-26
https://doi.org/10.22042/isecure.2017.0.0.5
E. Aerabi, M. Kaykha, M. Fazeli, A. Patooghy, A. Akbari
Abstract Embedded systems are suggestive targets for code injection attacks in the recent years. Software protection mechanisms, and in general computers, are not usually applicable in embedded systems since they have limited resources like memory and process power. In this paper we investigate side channel characteristics of embedded systems and their applicability in code injection attack detection. The architectural simulation for execution time, power usage and temperature on benchmarks shows that these parameters disclose meaningful and distinguishable behaviours in case of attack.