FPGA Realization of Hodgkin-Huxley Neuronal Model.

One of the appealing cases of the neuromorphic research area is the implementation of biological neural networks. The current study offers Multiplierless Hodgkin-Huxley Model (MHHM). This modified model may reproduce various spiking behaviors, like the biological HH neurons, with high accuracy. The presented modified model, in comparison to the original HH model, due to its exact similarity to the original model, has more top performances in the case of FPGA saving and more achievable frequency (speed-up). In this approach, the proposed model has a 69 % saving in FPGA resources and also the maximum frequency of 85 MHz that is more than other similar works. In this modification, all spiking behaviors of the original model have been generated with low error calculations. To validate the MHHM neuron, this proposed model has been implemented on digital hardware FPGA. This approach demonstrates that the original HH model and the proposed model have high similarity in terms of higher performance and digital hardware cost reduction.

Digital Multiplierless Realization of Coupled Wilson Neuron Model.

The human brain is composed of 1011 neurons with a switching speed of about 1 ms. Studying spiking neural networks, including the modeling, simulation, and implementation of the biological neuron models, helps us to learn about the brain and the related diseases, or to design more efficient bio-mimic processors and smarter robots. Such applications have made this part of neuromorphic research works very popular. In this paper, the Wilson neuron model has been implemented as an approximation of the Hodgkin-Huxley biological model that is adjusted for the efficient digital realization on the platforms. Results show that the proposed model can adequately reproduce neuron dynamical behaviors. The hardware implementation on the field-programmable gate array (FPGA) shows that our modifications on the Wilson original model imitate the biological behavior of neurons, besides using feasibility, targeting a low cost and high efficiency. The modifications raised a 15% speed-up compared with the original model. The mean normalized root-mean-square error, root-mean-square error, and the mean absolute error parameters are 6.43, 0.44, and 0.31, respectively.