Space-and-Cost-Efficient Neural Control /Sensory Element Using an Analog FPGA

As neural networks are applied to control and sensory, software neuron models cannot sometimes fulfill speed requirement as well as simple add-and-sigmoid is not enough for functionality. This paper proposes a small and inexpensive hardware neuron based on FitzHugh-Nagumo model. By making full use of chip property and maximum circuit packing through placement search, it surpasses the previous implementation by factors of 4 for space and 15 for cost.

[1]  Roger D. Quinn,et al.  A Dynamic Neural Network Designed Using Analytical Methods Produces Dynamic Control Properties Similar to an Analogous Classical Controller , 2019, IEEE Control Systems Letters.

[2]  Sina Balkir,et al.  Bio-inspired compact cell circuit for reaction-diffusion systems , 2005, IEEE Transactions on Circuits and Systems II: Express Briefs.

[3]  Ruixin Wang,et al.  Adaptive integratable hardware realization of analog neural networks for nonlinear system , 2015, 2015 IEEE 13th International Conference on Industrial Informatics (INDIN).

[4]  Yong-Bin Kim,et al.  Circuit implementation of FitzHugh-Nagumo neuron model using Field Programmable Analog Arrays , 2007, 2007 50th Midwest Symposium on Circuits and Systems.

[5]  Costas Psychalinos,et al.  Low-Voltage Low-Power Integrable CMOS Circuit Implementation of Integer- and Fractional–Order FitzHugh–Nagumo Neuron Model , 2019, IEEE Transactions on Neural Networks and Learning Systems.

[6]  J. NAGUMOt,et al.  An Active Pulse Transmission Line Simulating Nerve Axon , 2006 .

[7]  Pham Thuong Cat,et al.  Real-time Image Processing by Cellular Neural Network Using Reaction-Diffusion Model , 2009, 2009 International Conference on Knowledge and Systems Engineering.

[8]  R. FitzHugh.  Impulses and Physiological States in Theoretical Models of Nerve Membrane. , 1961, Biophysical journal.

[9]  Xiaofeng Liao,et al.  Synchronization and chaos in coupled memristor-based FitzHugh-Nagumo circuits with memristor synapse , 2017 .

[10]  T. Hidvegi,et al.  A Programmable Digital Cellular Neural Network Processing On- and Off-Chip Sensory Information , 2006, 2006 10th International Workshop on Cellular Neural Networks and Their Applications.

[11]  Simha Sethumadhavan,et al.  Analog Computing in a Modern Context: A Linear Algebra Accelerator Case Study , 2017, IEEE Micro.