False Glitch Cells as a Countermeasure Against Power Analysis Attacks in Cryptographic Circuits

Cryptographic systems can potentially suffer from various attacks which are a consequence of vulnearability of side channel information to analysis attacks. The nature of digital circuits is prone to electromagnetic and power analysis (EMA and PA) attacks. Interconnects in cryptographic circuits emanate information through electromagnetic signals which can be exploited using EMA. Complementary Metal Oxide Semiconductor (CMOS) circuits are vulnearable to leak of computational information through leakage and dynamic power consumption. Dynamic power consumption of cryptographic engines is a rich source to sink in the data under process. The hardware implementation of cryptographic algorithms undermine the security of encrypted system against non brute-force-attacks. The attackers collect and analyse the leakage information of cryptographic circuits to detect the secrete key. The parameters sensitivity of the device and circuits make the cryptographic engines vulnearable to fault injection and inherent inconsistency of the circuits. Dynamic and static power comprise two major components of power consumption of digital circuits. Both power components vary with data under process. This exhibits power consumption has data dependency in digital circuits. To make noisy power profile, various countermeasures have been proposed against Differential Power Analysis (DPA) attacks. In this research, false glitch cells is proposed to work as a countermeasure to generate random transitions and power spikes to scramble power profile. The mathematical foundation of security implications of false glitch cells against Correlation Power Analysis (CPA) attacks is investigated. Moreover, the security efficiency of utilizing false glitch cells is verified with different metrics.