Python-Based Circuit Design for Fundamental Building Blocks of Spiking Neural Network

Spiking neural networks (SNNs) are considered a crucial research direction to address the “storage wall” and “power challenges faced by traditional artificial intelligence computing. However, developing SNN chips based on CMOS (complementary metal oxide semiconductor) circuits remains challenge. Although memristor process technology is best alternative synapses, it still undergoing refinement. In this study, novel approach proposed that employs tools automatically generate HDL (hardware description language) code for constructing neuron after using Python describe models. Based approach, HR (Hindmash–Rose), LIF (leaky integrate-and-fire), IZ (Izhikevich) circuits, as well HP, EG (enhanced generalized), TB (the behavioral threshold bipolar) designed construct most basic connection of SNN: neuron–memristor–neuron circuit satisfies STDP (spike-timing-dependent-plasticity) learning rule. Through simulation experiments FPGA (field programmable gate array) prototype verification, confirmed suitable neurons in SNNs, while X variables model serve characteristic synaptic weights. The satisfy rule synapses SNNs. comparison previous works hardware spiking neurons, method needed fewer area resources creating models FPGA. components design method, resulting beneficial architectural exploration hardware–software co-design chips.