A Modified DPI-Neuron with Various Biological Firing Patterns

Neuromorphic VLSI is an important tool for investigating and implementing neural algorithms. The DPI-neuron analog implementation of neural network in silicon is the real-time conductance-based that has the low power consumption. This circuit implemented Exponential Integrated-and-Fire model that produces real time spiking behavior. In this paper we modify DPI-neuron circuit which can be produce different type of behavior biological neuron. It is suitable for applications in reconfigurable neuromorphic devices which implement biologically brain circuits. This circuit is simulated by HSPICE software in 0.35m CMOS technology with supply voltage 3.3Volt.

Keywords:

Neuromorphic , DPI neuron , low power , CMOS