Digital cellular implementation of Morris-Lecar neuron model
Digital cellular implementation of Morris-Lecar neuron model
复制标题
Morris-Lecar 神经元模型的数字细胞实现
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
S. Saeedi
中科院分区：
文献类型：
--
作者：
M. Gholami;S. Saeedi
详细的生物神经元模型，如Morris-Lecar，由于其方程中存在非线性函数和复杂运算，无法使用传统的基于欧拉的数字或模拟方法轻松实现。本研究提出一种有效的细胞为基础的数字架构，实现莫里斯-勒卡神经元模型。数字硬件后合成结果表明，该硬件模型能够再现生物模型的各种响应。所提出的架构是不依赖于方程的复杂性，并不适用于函数逼近方法来提供可实现的方程。这意味着所有其他详细的神经元模型也可以通过这种结构来实现。所提出的硬件模型的高度可编程性也使其能够应用于嵌入式神经形态硬件和实时应用。
The detailed biological neuron models such as Morris-Lecar cannot be easily implemented using conventional digital or analog Euler-based methods due to the presence of nonlinear functions and complex operations in their equations. This study presents an efficient cellular-based digital architecture for implementing Morris-Lecar neuron model. Digital hardware post synthesis results show that this hardware model is able to reproduce various responses of the biological model. The proposed architecture is not dependent on the complexity of the equations, and applies no function approximation method to deliver implementable equations. This implies that all other detailed neuron models can also be implemented by this structure. High programmability of the proposed hardware model also enables it to be applied to embedding neuromorphic hardware and real-time applications.