Abstract—This article presents an innovative reconfigurable parametric model of a digital Spiking Neuron (SN). The model is based on the classical Leaky Integrate and Fire (LIF) model with some unique modifications, allowing neuron configuration using seven different leak modes and three activation functions with dynamic threshold setting. Efficient hardware implementation of the proposed spiking neuron significantly reduces the area and power cost. The SN model is implementable requiring only 700 ASIC 2-inputs gates and is denser than IBM SN which is composed of 1272 gates. The proposed spiking neuron model can be expanded to support a larger recurrent neural network and is efficiently applied to audio applications. Performance evaluation has been carried out through a simple voice activation system design using 1000 SNs, demonstrating ultra-low power consumption of only 20uW and consuming an area of 0.03 mm2 using 28nm technology. Simulations of the proposed digital spiking neuron also demonstrate its ability to accurately replicate the behaviors of a biological neuron model.
Index Terms—Spiking neuron, digital neuron, recurrent neural network, LIF model, LSTM, low power, voice activation.
The author are with the Ben-Gurion University, Israel (e-mail: bensimmo@post.bgu.ac.il, shlomog@bgu.ac.il, benshimo@bgu.ac.il, Moshe.Haiut@dspg.com).
[PDF]
Cite: Moshe Bensimon, Shlomo Greenberg, Yehuda Ben-Shimol, and Moshe Haiut, "A New Digital Low Power Spiking Neuron," International Journal of Future Computer and Communication vol. 8, no.1, pp. 24-28, 2019.
