Parametric Analysis of Spiking Neurons in 16 nm Fin Field‐Effect Transistor Technology

The saturation of traditional computing hardware at criticalthermal densities necessitates a paradigm shift toward energy‐efficient, brain‐inspired computing. This study explores the design of three distinct complementary metal‐oxide‐semiconductor‐based spiking neuron topologies: Leaky Integrate‐and‐Fire (LIF), Morris–Lecar (ML), and Axon‐Hillock (AH), implemented in a 16 nm fin field‐Effect transistor technology. Through extensive parametric optimization of supply voltage and internal capacitance, we characterize the fundamental trade‐offs between energy consumption, maximum firing frequency, and functional robustness. Our results demonstrate that while the AH circuit achieves the highest throughput with spiking frequencies exceeding 1.51 GHz (at 0.4 V, 0.01 fF), it exhibits functional fragility, failing to register coincident spikes with output voltages remaining below 1 mV. In contrast, the ML model captures complex biological dynamics, specifically demonstrating rebound spiking where inhibitory loading increased firing frequency by 25% (from 0.63 to 0.79 GHz at 0.7 V, 1 fF). The LIF model emerges as the optimal candidate for general‐purpose neuromorphic engineering, maintaining a stable firing rate of ∼1.0 GHz (at 0.7 V, 1 fF) across the inhibition spectrum while providing robust coincidence detection. These findings provide a quantitative framework for matching neuron topologies with specific architectural requirements in next‐generation high‐density neuromorphic systems.