TY - JOUR

T1 - Modeling and Evaluating Errors Due to Random Clock Shifts in Quantum-Dot Cellular Automata Circuits

AU - Karim, F.

AU - Ottavi, Marco

AU - Hashempour, H.

AU - Vankamamidi, Vamsi

AU - Walus, K.

AU - Ivanov, Andre

AU - Lombardi, F.

PY - 2009

Y1 - 2009

N2 - This paper analyzes the effect of random phase shifts in the underlying clock signals on the operation of several basic Quantum-dot Cellular Automata (QCA) building blocks. Such phase shifts can result from manufacturing variations or from uneven path lengths in the clocking network. We perform numerical simulations of basic building blocks using two different simulation engines available in the QCADesigner tool. We assume that the phase shifts are characterized by a Gaussian distribution with a mean value of iπ2, where i is the clock number and a standard deviation, σ, which is varied in each simulation. Our results indicate that the sensitivity of building blocks to phase shifts depends primarily on the layout while the reliability of all building blocks starts to drop once the standard deviation, σ exceeds 4°. A full adder was simulated to analyze the operation of a circuit featuring a combination of the building blocks considered here. Results are consistent with expectations and demonstrate that the carry output of the full adder is better able to withstand the phase shifts in the clocking network than the Sum output which features a larger combination of the simulated building blocks.

AB - This paper analyzes the effect of random phase shifts in the underlying clock signals on the operation of several basic Quantum-dot Cellular Automata (QCA) building blocks. Such phase shifts can result from manufacturing variations or from uneven path lengths in the clocking network. We perform numerical simulations of basic building blocks using two different simulation engines available in the QCADesigner tool. We assume that the phase shifts are characterized by a Gaussian distribution with a mean value of iπ2, where i is the clock number and a standard deviation, σ, which is varied in each simulation. Our results indicate that the sensitivity of building blocks to phase shifts depends primarily on the layout while the reliability of all building blocks starts to drop once the standard deviation, σ exceeds 4°. A full adder was simulated to analyze the operation of a circuit featuring a combination of the building blocks considered here. Results are consistent with expectations and demonstrate that the carry output of the full adder is better able to withstand the phase shifts in the clocking network than the Sum output which features a larger combination of the simulated building blocks.

U2 - 10.1007/s10836-008-5088-9

DO - 10.1007/s10836-008-5088-9

M3 - Article

VL - 25

SP - 55

EP - 66

JO - Journal of electronic testing

JF - Journal of electronic testing

SN - 0923-8174

ER -