An Artificial Spiking Afferent Neuron System Achieved by 1M1S for Neuromorphic Computing

A calibratable sensory neuron based on epitaxial VO2 for spike-based neuromorphic multisensory system

“…Given that our pressure sensor is 2 cm in diameter, the calculated sensitivity to pressure is 47.67 kHz/kPa, which is slightly lower than the 60.8 kHz/kPa reported in ref. 30 . This can be improved by increasing the sensitivity of the pressure sensor itself.…”

“…24 schematically depicts the comparison between neuromorphic perception system based on silicon circuits and our approach. In traditional silicon-based circuits, in order to sense physical signals a large number of ADCs (analog-to-digital converters) are necessary besides the sensors, which are very costly in area and energy consumption, and when the subsequent information processing is in spike-based neuromorphic computing systems, a large number of additional VSCs (voltage-to-spike converters) will be required 30 to realize spike conversion, which also consume a large amount of area and energy, as shown in Supplementary Fig. 24a .…”

Ovonic threshold switching-based artificial afferent neurons for thermal in-sensor computing

“…8,9 Moreover, these synthetically perceptive artificial afferent neurons can be applied to abundant scenarios at the edge, such as intelligent sensors, image recognizers and nociceptors, which provides potential for bio-electronic interfaces and the input of the spiking neural network (SNN). 10–16 However, the conventional CMOS-based artificial afferent neurons require intricate peripheral circuits including analog-to-digital converters (ADCs) and voltage-to-spike converters (VSCs), 17,18 resulting in inefficiencies in energy and overall area utilization. 19…”

Artificial Tactile Perception System Based on Spiking Tactile Neurons and Spiking Neural Networks

“…Neuromorphic tactile perception systems aim to emulate the efficient biological systems, offering parallel distributed sensory information processing, low energy consumption, and high fault tolerance. − A critical component in this context is the artificial tactile neuron, responsible for perceiving tactile signals and encoding them into electrical spikes for further processing. − However, many existing artificial tactile neurons rely on transistor-based oscillators for the analog-to-spike conversion, leading to increased hardware consumption …”

“…11−18 However, many existing artificial tactile neurons rely on transistor-based oscillators for the analog-tospike conversion, leading to increased hardware consumption. 14 To address these challenges, researchers have explored using volatile metal−insulator transition (MIT) memristors, known for their two-terminal structure and dynamic threshold switching characteristics, as promising candidates for constructing tactile neurons. 19−21 The integration of MIT memristors into artificial tactile neurons has shown promise in achieving low energy consumption and enabling parallel distributed processing.…”