Answers are grounded in this paper and cite their sources.
This research demonstrates gas classification with 99.6% accuracy in mobile devices, implying ultralow-power solutions may address existing limitations.
With increasing demand for gas sensors in mobile devices, research on developing an electronic nose (E-nose) is actively conducted. However, conventional E-nose systems based on von Neumann computing have encountered challenges such as high hardware costs and power consumption because of the necessity of hardware-intensive circuits and processors. This work implements low-power artificial olfactory neuron modules within a spiking neural network (SNN) to address this issue. The artificial olfactory neuron module is developed by connecting a GeSe-based ovonic threshold switch and a micro–light-emitting diode (μLED) platform–based semiconductor metal oxide gas sensor in series. The use of μLED gas sensors enables ultralow-power operation, resulting in substantially decreased power consumption. The artificial olfactory neuron module generates spike signals with low operation voltage, demonstrating energy efficiency and advanced performance. A real-time gas classification based on the SNN is feasibly conducted with an accuracy of 99.6%. Moreover, it is possible to classify different ingredients under humidity disturbance conditions through a hardware SNN.
No takes yet. Share an insight, caveat, or question.
Kang et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: