Related Experiment Video
Updated: May 20, 2026
05:19
Using Neuron Spiking Activity to Trigger Closed-Loop Stimuli in Neurophysiological Experiments
Published on: November 12, 2019
Hardware implementation of stochastic spiking neural networks
Josep L Rosselló1, Vincent Canals, Antoni Morro
1Physics Department, Universitat de les Illes Balears, Cra. Valldemossa km. 7.5, Palma de Mallorca, Balears, 07122, Spain.
International Journal of Neural Systems
|July 27, 2012
Summary
This study introduces a novel digital hardware implementation of Spiking Neural Networks (SNNs) that incorporates stochastic processes found in biological neurons. This bio-inspired approach enables high-speed signal filtering and complex equation solving using Field Programmable Gate Arrays.
Area of Science:
- Neuroscience and Computer Engineering
- Artificial Intelligence and Hardware Acceleration
Background:
- Spiking Neural Networks (SNNs) mimic biological neurons, offering superior computational power.
- Stochastic processes in biological neurons are crucial for their unique arithmetic capabilities.
Purpose of the Study:
- To present a simple, fully digital hardware implementation of spiking neurons that accounts for their probabilistic nature.
- To demonstrate the computational advantages of this bio-inspired hardware model.
Main Methods:
- Developed a fully digital hardware architecture for spiking neurons incorporating stochastic elements.
- Implemented feed-forward and recurrent network configurations on Field Programmable Gate Arrays (FPGAs).
Main Results:
- The digital hardware implementation successfully models the probabilistic nature of biological neurons.
- Networks built with these neurons achieved high-speed signal filtering.
- The networks demonstrated the ability to solve complex systems of linear equations.
Conclusions:
- The proposed digital hardware implementation of SNNs offers a scalable and efficient approach for advanced computational tasks.
- This bio-inspired, stochastic model provides a powerful platform for real-time signal processing and complex problem-solving in hardware.
