| |
Saber Moradi, Sunil Bhave, and Rajit Manohar
Designing analog sub-threshold neuromorphic circuits in deep sub-micron technologies e.g. 28 nm can be a daunting task due to the problem of excessive leakage current. We propose novel energy-efficient hybrid CMOS-nano electro-mechanical switches (NEMS) Leaky Integrate and Fire (LIF) neuron and synapse circuits and investigate the impact of NEM switches on the leakage power and overall energy consumption. We analyze the performance of biologically-inspired neuron circuit in terms of leakage power consumption and present new energy-efficient neural circuits that operate with biologically plausible firing rates. Our results show the proposed CMOS-NEMS neuron circuit is, on average, 35% more energy-efficient than its CMOS counterpart with same complexity in 28 nm process. Moreover, we discuss how NEM switches can be utilized to further improve the scalability of mixed-signal neuromorphic circuits.
|
|
