Spiking Neural Network Based Low-Power Radioisotope Identification using FPGA

this paper presents detailed methodology of a Spiking Neural Network (SNN) based low-power design for radioisotope identification. A low power cost of 72 m W has been achieved on FPGA with the inference accuracy of 100% at 10 cm test distance and 97% at 25 cm. The design verification and chip valida...

Full description

Saved in:
Bibliographic Details
Published in2020 27th IEEE International Conference on Electronics, Circuits and Systems (ICECS) pp. 1 - 4
Main Authors Huang, Xiaoyu, Jones, Edward, Zhang, Siru, Xie, Shouyu, Furber, Steve, Goulermas, Yannis, Marsden, Edward, Baistow, Ian, Mitra, Srinjoy, Hamilton, Alister
Format Conference Proceeding
LanguageEnglish
Published IEEE 23.11.2020
Subjects
event-based signal processing
Field programmable gate arrays
FPGA
Hardware
low power
Mathematical model
Monitoring
Neurons
Photonics
Radioactive materials
radioisotope identification
spiking neural networks
SpiNNaker
Online AccessGet full text

Cover

More Information
Summary:this paper presents detailed methodology of a Spiking Neural Network (SNN) based low-power design for radioisotope identification. A low power cost of 72 m W has been achieved on FPGA with the inference accuracy of 100% at 10 cm test distance and 97% at 25 cm. The design verification and chip validation methods are presented. It also discusses SNN simulation on SpiNNaker for rapid prototyping and various considerations specific to the application such as test distance, integration time and SNN hyperparameter selections.
DOI:10.1109/ICECS49266.2020.9294873