A CORDIC Based Digital Hardware For Adaptive Exponential Integrate and Fire Neuron

作者：Heidarpour Moslem; Ahmadi Arash^*; Rashidzadeh Rashid

来源：IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS, 2016, 63(11): 1986-1996.

DOI：10.1109/TCSI.2016.2598161

• Adaptive exponential integrate and fire (AdEx)
• biological neuron model
• CORDIC
• digital implementation
• neuromorphic
• spiking neural network

This paper presents a COordinate Rotation DIgital Computer (CORDIC) based Adaptive Exponential Integrate and Fire (AdEx) neuron for efficient large scale biological neural network implementation. The accuracy of the modified model is investigated by both calculating various errors and bifurcation analysis; both show that the proposed model foll ows the same signaling, dynamical behavior, and bifurcation pattern as the original model. Network behavior of the original and proposed models are also observed to be very much alike in following the same activity patterns. The model is hardware synthesized and implemented on FPGA as a proof of concept. Measurement results show that the proposed model can reproduce neuronal behaviors similar to the original model. Hardware device utilization and speed also confirm the efficiency of the realized hardware compared with previous works.

• 出版日期2016-11