Chip side channel attack noise reduction preprocessing method based on residual learning

Abstract

The invention relates to a chip side channel attack noise reduction preprocessing method based on residual learning, and belongs to the technical field of chip security detection. The invention provides a chip side channel attack noise reduction preprocessing method based on residual learning, and the method comprises the steps: learning a mapping relation between an original energy trace and noise through constructing a deep residual network, thereby generating any noise corresponding to an energy trace which accords with the characteristics of collection equipment, subtracting the generated noise from the original energy trace, and completing the noise reduction pretreatment step.

Description

technical field [0001] The invention belongs to the technical field of chip security detection, and in particular relates to a chip side channel attack noise reduction preprocessing method based on residual error learning. Background technique [0002] Cryptographic algorithms exist in information security systems in the form of software or hardware implementations, and these systems are usually deployed in an untrusted or even malicious environment. When any electronic device is running, it may leak physical state information closely related to its internal state (time difference, energy consumption, electromagnetic radiation, sound leakage, light radiation, heat radiation, etc.). Attackers can use the physical state information leaked when these devices are running to recover key information. This type of attack method is called a side channel attack. Side-channel attacks are passive and non-invasive. Attackers can detect and collect encrypted side information (typically ...

AI Technical Summary

Problems solved by technology

However, with the increasing integration of cryptographic chip circuits, the increasing types of security protection measures, and the in-depth application of multi-core pipeline technology, the types of noise in cryptographic chip information leakage are becoming more and more complex, and the negative impact of noise on side channel analysis is more serious. It is difficult to effectively reduce the adverse effects of noise with noise reduction methods

Take the filtering noise reduction algorithm as an example. On the one hand, this type of method focuses on the noise reduction of a single energy trace itself at the algorithm level, and it is difficult to use and learn the distribution characteristics of the noise on the entire data set; on the other hand, for different The implementation type of software and hardware cryptography. There is no suitable guidance for parameter selection in this method. The development of the technology strongly depends on the professional experience and trial and error time of the analysts, which leads to a severe limitation of its noise reduction effect, and sometimes even has the negative effect of introducing additional noise.

Images