Hamdioui, SaidTaouil, MottaqiallahAljuffri, Abdullah2024-03-062024-03-062024-02-27https://hdl.handle.net/20.500.14154/71593The security of electronic devices holds the greatest importance in the modern digital era, with one of the emerging challenges being the widespread occurrence of hardware attacks. The aforementioned attacks present a substantial risk to hardware devices, and it is of utmost importance to comprehend the potential detrimental effects they may cause. Side-channel attacks are a class of hardware attacks that exploit information unintentionally leaked by a device during its operation. These leaks manifest in various forms, including power consumption, time variations, and thermal dissipation. The fundamental danger posed by side-channel attacks is their ability to infer sensitive information from these unintended emissions. To address the heightened risks associated with side-channel attacks, this thesis focuses on three main research topics. Side Channel Analysis: Side-channel attacks can manifest in various forms, depending on the specific leakage channels employed. The present study primarily focused on the investigation of three distinct categories of leakage, as it is hypothesized that these specific forms of leakage present the greatest potential risks. The aim of the analysis is to identify the optimal channels for creating an assessment framework. The selected leakages for analysis cover power consumption, temporal variations, and thermal attacks. Power consumption measurements provide valuable insights into the behavior and execution patterns of algorithmic operations, facilitating the identification of specific operations that are particularly vulnerable to attacks. There are other types of leakages that are similar, such as electromagnetic emissions. However, it is important to note that power consumption demonstrates considerably lower levels of noise. The use of time variations in evaluating operations is subject to certain limitations due to the need to wait for a response. Nevertheless, one notable advantage of these systems is their ability to offer convenient remote access, facilitated by their software-based calculation capabilities. Despite its inherent noise, thermal monitoring is employed in nearly all devices as a means to prevent overheating. The ability to remotely access this monitoring system is facilitated through software. Consequently, a meticulous examination is necessary to identify potential modes of assault. Countermeasures: Cryptographic algorithms and other security primitives are the basic components of any cryptosystem. In their most optimized versions, these algorithms are frequently thought to be prone to side-channel attacks (SCAs), which necessitates the development of countermeasures. In this thesis, four countermeasures that have been developed are thoroughly analyzed. The countermeasures that were devised covered a wide range of algorithms, such as GIFT, RSA, and AES, and they were suitable for a variety of applications, including lightweight ones. The first countermeasure that has been developed makes use of an Advanced Encryption Standard (AES) implementation that is based on neural networks. This countermeasure's principal goal is to confuse the attacker by causing random fluctuations in power consumption. The second countermeasure is developed for asymmetric algorithms. This countermeasure's goal is to balance the leakage by making power consumption similar among all its executions. The goal of developing the third algorithm was to provide a countermeasure that is lightweight and tailored to symmetric algorithms. This countermeasure is based on the integration of balancing and randomization techniques. To ensure that the results of these two operations show balanced power behaviors in a random way, two instances of the SBOX operation are generated to complement each other. The fourth countermeasure involves the optimization of a widely known countermeasure named Domain-Oriented Masking (DOM) to adapt to lightweight applications. The countermeasure used in this research combines optimization techniques like resource sharing, module optimizations, and key-expansion bypassing. Pre-silicon Leakage Assessment: After recognizing the importance of mitigating side-channel leakages and developing various countermeasures, the subsequent phase entails establishing a framework for evaluating these vulnerabilities. In contrast to software vulnerabilities, which can be addressed through patching at any given time, the mitigation of hardware vulnerabilities necessitates expensive modifications to the physical hardware. Hence, it is essential to develop a leakage assessment framework that can effectively evaluate the system during the design phase. In this thesis, we present an innovative and pioneering methodology that relies on the application of Generative Neural Networks (GANs). The methodology described herein signifies a substantial advancement in the pursuit of enhanced security in the field of chip design. This framework demonstrates outstanding ability to rapidly produce traces that closely correspond to those obtained from computer-aided design (CAD) processes. As a result, it enables the efficient validation of numerous countermeasures within a realistic timeframe.160enside channel analysisPower attacksCountermeasuresLeakage Assessment FrameworkHardware SecurityThesis