A modified adaptive exponential integrate and fire neuron model for circuit implementation of spiking neural networks

Nowadays neuroscience is progressing to higher levels which have made it possible to have a better understanding of the brain behavior. In this scheme, spiking neural network has a great potential and have attracted much research interests. In this direction, one problem in simulations and implementations is speed and simplicity. In other hand, neuron models as building blocks of the neuronal systems have a vital role. One of the recently developed neuron models is called “Adaptive Exponential Integrate and Fire”. This model is a two dimensional system that can produce rich firing pattern. This paper proposes simplified models based on the Adaptive Exponential Integrate and Fire model. This modification simplifies the hardware implementation, increases speed and demonstrates similar dynamic behavior. These models can be used for both of analog and digital implementations. This paper can be a step in the neural network simulation and implementation as large as the brain scale.