Mitigating the Impact of Variability in NCFET-based Coupled-Oscillator Networks Applications

Coupled oscillators are attracting increasing interest because of their potential to perform computation efficiently, enabling new applications in computing and information processing. Coupled nano-oscillator implementations using emerging devices have arisen, but the immaturity of these technologies has allowed only simple experimental demonstrations. The potential of Negative Capacitance FET (NCFET) for such applications has recently been recognized, which is a step towards the physical realization given their ease of co-integration with commercial CMOS technologies. However, the design of circuits using these devices can be seriously compromised by the variability inherent in them. In this paper, we will highlight this problem through the design of an oscillatory neural network for pattern recognition applications. We propose the application of subharmonic injection mechanisms to mitigate the impact of NCFET transistor variability and present results showing that the performance of these circuits improves significantly.