Numerical Simulation And Hardware Implementation Of Neurons Action Potential Based On Hodgkin-Huxley Model

The nervous system is a complex nonlinear system. Neuron is the basic structural and functional unit of the nervous system. Action potential is the medium which is used to transmit information between neurons in the neuronal network and it plays an important role in neural information processing. In this dissertation, only single neuron is considered. The Hodgkin-Huxley(HH) model which is the most realistic neuron models on the electrophysiological characteristic description is selected as the neuronal mathematical model, and the numerical simulation and hardware implementation of the HH model neuron are all investigated. The following theoretical problems are discussed specifically:The first part provides a brief introduction of neuron and neuronal mathematical models. Structure of neuron, resting potential(RP) and action potential(AP) of neuron and the ionic mechanisms of AP are described in detail. At the same time, the history of HH model, the significance of HH model and the other common neuronal mathematical model are also described. The results show that the HH model is the most realistic neuron model on the electrophysiological characteristic description of neuron.Numerical simulation of the HH model neuron is studied in the second part. The characteristics of single neuron, such as threshold effect, refractory period effect of neurons are mainly studied. The numerical simulations of neuron under different types of current stimulation including square and sine waves are also studied. The results show that the HH neuron model could simulate threshold effect and refractory period effect of neuron perfectly, and show different firing characteristics in different ion channel parameters or under different types of current stimulation. The firing characteristics could be explained by the threshold effect and refractory period effect of neurons.Based on the HH neuron model and the DSP Builder technology, a single HH neuron model hardware implementation is completed in Field Programmable Gate Array(FPGA) in this dissertation. The action potential response characteristics of the neuron that implemented in FPGA are analyzed under different types of current stimulation, and the results of hardware implementation and numerical simulation compared to verify the correctness of this method. In order to realize neural network, two coupled neurons with hardware implementation is also investigated in this dissertation. The results show that FPGA could be used to complete hardware realization of HH neuron model. The action potential response from FPGA is highly agreement with the result of numerical simulation, and it is possible to realize small scale neuronal network via FPGA arrays.