An implementation method of pulsed neuron model for FPGA devices

摘要

We have been studied about the pulsed neural networks which are suitable for temporal information processing. The pulsed neural networks have very simple structure and don't need a multiplier for implementation. In this article, we proposed an implementation method of a pulsed neural network for the FPGA devices. By apply this method, we can implement many pulsed neuron model in a FPGA device. We implemented pulsed neuron modules for sound localization and went through the real-time processing of them on a FPGA device.