Multiple Missing Cell Defect Modeling for QCA Devices

Considering the limitations of CMOS technology, the Quantum-dot Cellular Automata (QCA) is emerging as one of the alternatives for Integrated Circuit (IC) Technology. A lot of work is being carried out for design, fabrication and testing of QCA circuits. In this paper, we have worked on defect analysis, fault models development and deriving various properties for QCA Majority Voter (MV) to effectively generate the test patterns for QCA circuits. It has been shown that unlike CMOS technology, single missing cell consideration is not enough for QCA technology. We have presented that the Multiple Missing Cell (MMC) defect, which is very natural at nanoscale, causes the sizable difference in functionality compared to Single Missing Cell consideration described in literature, and hence, must be considered while test generation. The proposed MMC is supported by exhaustive simulation results as well as kink energy based mathematical analysis. Further, Verilog fault models are proposed which can be used for the functional, timing verification and activation of faults caused by MMC defect. The effect of MMC on output is analyzed in stand-alone MV as well as when MV is a part of circuit. At the end, we have proposed the test properties of MV when being used as MV itself, as AND gate or OR gate. These properties may be further helpful in development of test generation algorithms.