Therefore, to ensure the accuracy of model translation, the translated model of DFF must fulfill all the above-mentioned properties, design constraints and characteristics. Similarly, the dynamic power, functionality and other characteristics are well defined in the literature.
7, CLK-to-Q delay of a DFF is equivalent to the combined delay of NAND1 and NAND3, hold time of a DFF is equivalent to the inverters, delay and the setup time of DFF is equivalent to the combined delay of NAND3 and NAND4. These attacks can activate their respective payload even if they are not directly affecting the cryptographic components. 3, the control signals of HT-1, 2, 4 and 5 can be used for DoS attacks, as a kill switch or for controlling the computation and communication. Moreover, most of these approaches are focused on the cryptographic modules but other modules can also be affected to launch Denial-of-Service attacks (DoS) or to disable the system. Symbolic side-channel parameters of these approaches makes the translated model deterministic, thus overshadows the uncertain behavior due to process variations. These SHTs remain undetected and inactive during the testing phase and even after deployment using external, internal or time-based triggers, they can initiate any of the drastic payloads as mentioned in Fig. Therefore, they only target the security vulnerabilities against Active Hardware Trojans and may not be able to analyze the security vulnerabilities due to Stealthy Hardware Trojans (SHT). These approaches provide the symbolic execution or fixed approximation of side-channel parameters. To overcome this issue, several approaches for performance modeling have been proposed, which model the side-channel parameters, i.e., power, propagation delay and temperature, and analyze the vulnerabilities based on temporal properties and property specification language. On the other hand, functional and behavioral model checking based approaches provide the comprehensive vulnerability analysis and cover all possible input test cases but they inherently pose the state-space explosion problem for complex systems. Therefore, they are unable to incorporate the parametric behavior, especially, leakage power, and effects of process variations, which limits their scope to Active Hardware Trojans (AHTs, that remain active even under the normal operation). However, these techniques deploy symbolic execution or fixed approximation of side-channel parameters. To address these issues, mathematical modeling and formal verification, (i.e., SAT solvers and model checking), based approaches provides the completeness and accuracy to some extent. These model-based simulations cannot guarantee complete coverage because of the computational constraints (energy and memory) and floating point inaccuracies. These techniques analyze the equivalent behavioral, functional and performance model against the design constraints and functionality characteristics. To address the above problem of high demand on the resources and the time of analysis during the design phase, different analytical techniques for vulnerability assessment against potential attacks have been developed.