The ISC International Journal of Information Security

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.

Abstract Most of the current research on static analysis of Android applications for security vetting either work on Java source code or the Dalvik bytecode. Nevertheless, Android allows developers to use C or C++ code in their programs that is compiled into various binary architectures. Moreover, Java and the native code components (C or C++) can collaborate with each other using Java Native Interface. Recent research shows that native codes are frequently used in both benign and malicious Android applications. Most of the present Android static analysis tools avert considering native codes in their analysis and applied trivial models for their data-flow analysis. As we know only the open source JN-SAF tool has tried to solve this issue statically. However, there are still challenges like libC functions and multi-threading in native codes that we want to address in this work. We presented SANT as an extension of JN-SAF for supporting Static Analysis of Native Threads. We considered modeling libC functions in our data-flow analysis to have a more precise analysis when dealing with security vetting of native codes. We also used control flow and data dependence graphs in SANT to handle multiple concurrent threads and find implicit data-flow between them. Our experiments show that the conducted improvements outperforms JN-SAF in real-world benchmark applications.