New Model For Wilson and Morris–Lecar Neuron Models: Validation and Digital Implementation on FPGA

Abstract

The neuron can be referred to as the primary cell of the nervous system, responsible for transmitting messages through electrical signals between neurons or other cells. To comprehend the intricate behavior of neuron performance, a set of differential equations incorporating non-linear functions is necessary. This research introduces a novel method to improve the digital implementation of Wilson and Morris-Lecar neuron models, offering multiple benefits such as decreased hardware needs, enhanced processing speed and accuracy, and reduced implementation expenses. The proposed method simplifies the original model by converting all its differential equations into a single trigonometric function. This transformation greatly reduces computational complexity by eliminating the need for multipliers, resulting in a concise set of mathematical expressions. The digital implementation of this innovative approach can be efficiently achieved using the COordinate Rotation DIgital Computer (CORDIC) algorithm, which avoids the need for complex mathematical operations. To showcase the effectiveness of this approach, the suggested model is synthesized and implemented on a Field-Programmable Gate Array (FPGA) with successful results. The implementation outcomes reveal a significant improvement in operational frequency, approximately 3.28 and 4.68 times (for Wilson and Morris-Lecar, respectively) higher than the original model.

Keywords

• Moris-Lecar
• Wilson
• FPGA
• digital implementation
• neuron model