An Investigation on Spiking Neural Networks Based on the Izhikevich Neuronal Model: Spiking Processing and Hardware Approach

The main required organ of the biological system is Central Nervous System (CNS), which can influence other basic organs in human body. elements this important are neurons, synapses, and glias (such as astrocytes, highest percentage brain). Investigating, modeling, simulation, hardware implementation (realization) different parts CNS case achieving a comprehensive neuronal that capable emulating all aspects real nervous system. This paper uses neuron model called Izhikevich to achieve high copy primary block, regenerating behaviors brain. proposed approach regenerate similarity degree performances. new based on Look-Up Table (LUT) modeling mathematical neuromorphic systems, be realized correlation with original model. procedure considered three cases: 100 points LUT 1000 10,000 modeling. Indeed, by removing high-cost functions model, presented implemented low-error, high-speed, low-area resources state comparison To test validate final hardware, digital FPGA board (Xilinx Virtex-II board) used. Digital synthesis illustrates our follow high-speed (more than model), increase efficiency, also reduce overhead costs. Implementation results show overall saving 84.30% higher frequency about 264 MHz, significantly 28 MHz.