A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures
Developing dedicated mixed-signal neuromorphic computing systems optimized for real-time sensory-processing in extreme edge-computing applications requires time-consuming design, fabrication, and deployment of full-custom neuromorphic processors. To ensure that initial prototyping efforts exploring...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | Neuromorphic Computing and Engineering |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/2634-4386/adef77 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849708889748013056 |
|---|---|
| author | Fernando M Quintana Maryada Pedro L Galindo Elisa Donati Giacomo Indiveri Fernando Perez-Peña |
| author_facet | Fernando M Quintana Maryada Pedro L Galindo Elisa Donati Giacomo Indiveri Fernando Perez-Peña |
| author_sort | Fernando M Quintana |
| collection | DOAJ |
| description | Developing dedicated mixed-signal neuromorphic computing systems optimized for real-time sensory-processing in extreme edge-computing applications requires time-consuming design, fabrication, and deployment of full-custom neuromorphic processors. To ensure that initial prototyping efforts exploring the properties of different network architectures and parameter settings lead to realistic results, it is important to use simulation frameworks that match as best as possible the properties of the final hardware. This is particularly challenging for neuromorphic hardware platforms made using mixed-signal analog/digital circuits, due to the variability and noise sensitivity of their components. In this paper, we address this challenge by developing a software spiking neural network simulator explicitly designed to account for the properties of mixed-signal neuromorphic circuits, including device mismatch variability. The simulator, called A R ealisti c Simulation Framework for A nalog/Digital N euromorphic A rchitectures, is designed to reproduce the dynamics of mixed-signal synapse and neuron electronic circuits with autogradient differentiation for parameter optimization and GPU acceleration. We demonstrate the effectiveness of this approach by matching software simulation results with measurements made from an existing neuromorphic processor. We show how the results obtained provide a reliable estimate of the behavior of the spiking neural network trained in software, once deployed in hardware. This framework enables the development and innovation of new learning rules and processing architectures in neuromorphic embedded systems. |
| format | Article |
| id | doaj-art-0eacf99202bf4bc584b10cf970e84063 |
| institution | DOAJ |
| issn | 2634-4386 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Neuromorphic Computing and Engineering |
| spelling | doaj-art-0eacf99202bf4bc584b10cf970e840632025-08-20T03:15:29ZengIOP PublishingNeuromorphic Computing and Engineering2634-43862025-01-015303400610.1088/2634-4386/adef77A Realistic Simulation Framework for Analog/Digital Neuromorphic ArchitecturesFernando M Quintana0https://orcid.org/0000-0001-5042-9399Maryada1https://orcid.org/0009-0009-9706-5989Pedro L Galindo2https://orcid.org/0000-0003-0892-8113Elisa Donati3https://orcid.org/0000-0002-8091-1298Giacomo Indiveri4https://orcid.org/0000-0002-7109-1689Fernando Perez-Peña5https://orcid.org/0000-0003-3586-2930Zernike Institute for Advanced Materials, University of Groningen , Groningen, The Netherlands; Groningen Cognitive Systems and Materials Center, University of Groningen , Groningen, The Netherlands; School of Engineering, University of Cádiz , Puerto Real, Cádiz, SpainInstitute of Neuroinformatics, University of Zurich and ETH Zurich , Zurich, SwitzerlandSchool of Engineering, University of Cádiz , Puerto Real, Cádiz, SpainInstitute of Neuroinformatics, University of Zurich and ETH Zurich , Zurich, SwitzerlandInstitute of Neuroinformatics, University of Zurich and ETH Zurich , Zurich, SwitzerlandSchool of Engineering, University of Cádiz , Puerto Real, Cádiz, SpainDeveloping dedicated mixed-signal neuromorphic computing systems optimized for real-time sensory-processing in extreme edge-computing applications requires time-consuming design, fabrication, and deployment of full-custom neuromorphic processors. To ensure that initial prototyping efforts exploring the properties of different network architectures and parameter settings lead to realistic results, it is important to use simulation frameworks that match as best as possible the properties of the final hardware. This is particularly challenging for neuromorphic hardware platforms made using mixed-signal analog/digital circuits, due to the variability and noise sensitivity of their components. In this paper, we address this challenge by developing a software spiking neural network simulator explicitly designed to account for the properties of mixed-signal neuromorphic circuits, including device mismatch variability. The simulator, called A R ealisti c Simulation Framework for A nalog/Digital N euromorphic A rchitectures, is designed to reproduce the dynamics of mixed-signal synapse and neuron electronic circuits with autogradient differentiation for parameter optimization and GPU acceleration. We demonstrate the effectiveness of this approach by matching software simulation results with measurements made from an existing neuromorphic processor. We show how the results obtained provide a reliable estimate of the behavior of the spiking neural network trained in software, once deployed in hardware. This framework enables the development and innovation of new learning rules and processing architectures in neuromorphic embedded systems.https://doi.org/10.1088/2634-4386/adef77SNNDPIneuromorphicPyTorchDYNAP-SE |
| spellingShingle | Fernando M Quintana Maryada Pedro L Galindo Elisa Donati Giacomo Indiveri Fernando Perez-Peña A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures Neuromorphic Computing and Engineering SNN DPI neuromorphic PyTorch DYNAP-SE |
| title | A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures |
| title_full | A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures |
| title_fullStr | A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures |
| title_full_unstemmed | A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures |
| title_short | A Realistic Simulation Framework for Analog/Digital Neuromorphic Architectures |
| title_sort | realistic simulation framework for analog digital neuromorphic architectures |
| topic | SNN DPI neuromorphic PyTorch DYNAP-SE |
| url | https://doi.org/10.1088/2634-4386/adef77 |
| work_keys_str_mv | AT fernandomquintana arealisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT maryada arealisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT pedrolgalindo arealisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT elisadonati arealisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT giacomoindiveri arealisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT fernandoperezpena arealisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT fernandomquintana realisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT maryada realisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT pedrolgalindo realisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT elisadonati realisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT giacomoindiveri realisticsimulationframeworkforanalogdigitalneuromorphicarchitectures AT fernandoperezpena realisticsimulationframeworkforanalogdigitalneuromorphicarchitectures |