Wei M.-LAmrouch HSung C.-LLue H.-TYang C.-LWang K.-CLu C.-Y.CHIA-LIN YANG2021-09-022021-09-02202115417026https://www.scopus.com/inward/record.uri?eid=2-s2.0-85105599032&doi=10.1109%2fIRPS46558.2021.9405141&partnerID=40&md5=51c329cc197e721fd3be551eb07a5cb7https://scholars.lib.ntu.edu.tw/handle/123456789/581288A spiking neural network (SNN) with non-volatile memory synapses can facilitate an ideal analog approach with low power consumption for intelligent applications. However, reliability issues would arise due to the characteristics of Non-Volatile Memory synapses operating under the limited size of a neural circuit component called membrane capacitor, resulting in low precision MAC. Through a simulation study, we identified several important criteria of the memory synapses that affect performance of the SNN for the membrane capacitor size of 1pF. (1) The required ON-OFF ratio of synapse needs to be >1000 to preserve classification accuracy. (2) Low on-current Ion (<10uA) is preferred for low power. (3) The variation and error of Ion should be lower than +/-10% of mean value. ? 2021 IEEE.Low power electronics; Nonvolatile storage; Reliability; Brain-inspired computing; Classification accuracy; Intelligent applications; Low-power consumption; Membrane capacitors; Non-volatile memory; Spiking neural network(SNN); Spiking neural networks; Neural networks[SDGs]SDG7conference paper10.1109/IRPS46558.2021.94051412-s2.0-85105599032