Asynchronous Cellular Automaton Based Modeling of Nonlinear Dynamics of Neuron

摘要

A modeling approach of nonlinear dynamics of neurons by an asynchronous cellular automaton is introduced. It is shown that an asynchronous cellular automaton neuron model can realize not only typical nonlinear response characteristics of neurons but also their underlying occurrence mechanisms (i.e., bifurcation scenarios). The model can be implemented as an asynchronous sequential logic circuit, whose control parameter is the pattern of wires that can be dynamically updated in a dynamic reconfigurable FPGA. An on-FPGA learning algorithm (i.e., on-FPGA rewiring algorithm) is presented and is used to tune the model so that it reproduces nonlinear response characteristics of a neuron.