A neuromodulable CMOS neuron

This project presents the first integrated circuit I designed during my PhD: an analog silicon neuron implemented in a 180 nm CMOS technology.

The neuron operates in the subthreshold regime, reproducing biologically-inspired spiking dynamics with extremely low energy consumption. Its main innovation is the addition of neuromodulation, allowing its behavior to be continuously tuned through dedicated analog control signals. This makes it possible to adapt the neuron’s dynamics in real time without modifying the circuit itself.

The design combines compactness, configurability, and biological realism, and served as the foundation for much of my subsequent work on adaptive and reliable neuromorphic hardware.

References

2025

Preprint

A Neuromodulable Current-Mode Silicon Neuron for Robust and Adaptive Neuromorphic Systems

Loris Mendolia, Chenxi Wen, Elisabetta Chicca, Giacomo Indiveri, Rodolphe Sepulchre, Jean-Michel Redouté, and Alessio Franci

2025

Bib HTML PDF

@misc{mendolia_neuromodulable_2025,
  title = {A Neuromodulable Current-Mode Silicon Neuron for Robust and Adaptive Neuromorphic Systems},
  author = {Mendolia, Loris and Wen, Chenxi and Chicca, Elisabetta and Indiveri, Giacomo and Sepulchre, Rodolphe and Redouté, Jean-Michel and Franci, Alessio},
  year = {2025},
  archiveprefix = {arXiv},
  primaryclass = {eess.SY},
  url = {https://arxiv.org/abs/2512.01133},
}

This page is under construction. More details and results will be added soon. In the meantime, you can check out the related publication (Mendolia et al., 2025)

References

2025