基于现场可编程门阵列的Hodgkin-Huxley模型神经元动作电位的数值模拟功能实现

发布时间：2018-02-12 09:55

本文关键词： Hodgkin-Huxley模型神经元现场可编程门阵列动作电位　出处：《生物医学工程学杂志》2015年06期 　论文类型：期刊论文

【摘要】：神经元是构成生物体神经系统的基本单位。在神经元的电生理特性描述上,Hodgkin-Huxley(HH)模型最接近于生物学实际。硬件实现神经元可为临床脊髓损伤治疗、仿生学、人工智能等方面提供新的研究思路。本文以HH模型神经元为研究对象,采用DSP Builder技术,在现场可编程门阵列(FPGA)上完成了单个HH模型神经元的硬件实现。对FPGA实现的神经元施加了不同类型的电流刺激,分析研究其动作电位响应特性,并与相同参数刺激下的数值模拟结果进行了相关系数的计算分析。结果表明,FPGA实现的神经元动作电位响应与数值模拟结果高度一致,为后续通过硬件构建神经元网络奠定了基础。
[Abstract]:The model of Hodgkin-Huxley HHH is the closest to the biological reality in describing the electrophysiological characteristics of neurons. The hardware realization of neurons can be used for the treatment of clinical spinal cord injury and bionics. In this paper, we take HH model neuron as the research object, adopt DSP Builder technology, The hardware implementation of a single HH model neuron has been completed on FPGA. Different types of current stimulation have been applied to the neurons realized by FPGA, and the action potential response characteristics of the neurons have been analyzed and studied. The correlation coefficient is calculated and analyzed with the results of numerical simulation under the same parameters. The results show that the response of neuron action potential realized by FPGA is in good agreement with the result of numerical simulation. It lays a foundation for the subsequent construction of neural network by hardware.
【作者单位】：兰州交通大学自动化与电气工程学院;兰州交通大学光电技术与智能控制教育部重点实验室;
【基金】：国家自然科学基金资助项目(51267010,51567015) 甘肃省杰出青年基金资助项目(1308RJDA013)
【分类号】：R338

【相似文献】