Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.00030 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909006588542976 |
|---|---|
| author | Nijsink, Wim Forlin, Bruno Endres Yousefzadeh, Amirreza Ottavi, Marco |
| author_facet | Nijsink, Wim Forlin, Bruno Endres Yousefzadeh, Amirreza Ottavi, Marco |
| contents | Neuromorphic, or spiking, processors are increasingly being considered for use in harsh, radiation-prone environments such as space and avionics, where energy efficiency and graceful degradation are essential. In this study, we propose and experimentally validate a radiation-testing methodology specifically designed for neuromorphic processors that employ on-chip synaptic plasticity. We map the open-source ODIN SNN processor with Spike-Dependent Synaptic Plasticity (SDSP) onto the FPGA and expose it to a high-energy neutron beam while continuously monitoring MNIST classification accuracy and recording the synaptic state. From these measurements, we extract SEU cross-sections for ODIN's synaptic memory and develop a calibrated fault model to inform a complementary fault-injection campaign. By comparing inference-only and online-learning configurations, we demonstrate that enabling SDSP can significantly extend the time to application-level failure and enable partial recovery from accumulated bit flips, with modest hardware overhead. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_00030 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Shooting Neutrons at Neurons: Radiation Testing of a Spiking Neural Network on Flash-Based FPGAs Nijsink, Wim Forlin, Bruno Endres Yousefzadeh, Amirreza Ottavi, Marco Hardware Architecture Systems and Control Neuromorphic, or spiking, processors are increasingly being considered for use in harsh, radiation-prone environments such as space and avionics, where energy efficiency and graceful degradation are essential. In this study, we propose and experimentally validate a radiation-testing methodology specifically designed for neuromorphic processors that employ on-chip synaptic plasticity. We map the open-source ODIN SNN processor with Spike-Dependent Synaptic Plasticity (SDSP) onto the FPGA and expose it to a high-energy neutron beam while continuously monitoring MNIST classification accuracy and recording the synaptic state. From these measurements, we extract SEU cross-sections for ODIN's synaptic memory and develop a calibrated fault model to inform a complementary fault-injection campaign. By comparing inference-only and online-learning configurations, we demonstrate that enabling SDSP can significantly extend the time to application-level failure and enable partial recovery from accumulated bit flips, with modest hardware overhead. |
| title | Shooting Neutrons at Neurons: Radiation Testing of a Spiking Neural Network on Flash-Based FPGAs |
| topic | Hardware Architecture Systems and Control |
| url | https://arxiv.org/abs/2605.00030 |