Delay-based reservoir computing using Mackey–Glass oscillator and Arduino board for edge intelligence applications
Source abstract
Abstract
This article introduces the experimental demonstration of the Mackey–Glass oscillator (MGO)/Arduino-based reservoir computing system as a novel versatile platform for several applications. Performance evaluations conducted on benchmark prediction tasks demonstrate the system’s capabilities with exceptional normalized mean square error (NMSE) values of up to 0.050 [log10(NMSE) ≃ −1.29] for Santa Fe and 0.0034 [log10(NMSE) ≃ −2.46] for electrocardiogram tasks. In addition, we achieve outstanding classification accuracy of up to 96.67% in the chaos recognition task. Our MGO/Arduino-based reservoir computing approach offers many advantages, such as cheapness, affordability, accessibility, and versatility, positioning it as a valuable and efficient solution in advancing neuromorphic computing for edge intelligence applications.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Neural Networks and Reservoir Computing · Advanced Memory and Neural Computing · Neural dynamics and brain function
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Research Unit of Condensed Matter, Electronics and Signal Processing, Department of Physics, University of Dschang 1 , P.O. Box 67, Dschang, Cameroon
Laboratoire d’Automatique et d’Informatique Appliquée, Department of Electrical Engeneering, IUT-FV Bandjoun, University of Dschang 2 , P.O. Box 134, Bandjoun, Cameroon
References (19)
Show source-provided references
1. A survey of deep learning: Platforms, applications and emerging research trends10.1109/access.2018.2830661 ↗
2. Cyber-physical systems architectures for industrial internet of things applications in industry 4.0: A literature review10.1016/j.jmsy.2020.11.017 ↗
3. A survey of the recent architectures of deep convolutional neural networks10.1007/s10462-020-09825-6 ↗
4. Photonic neuromorphic information processing and reservoir computing10.1063/1.5129762 ↗
5. Scaling for edge inference of deep neural networks10.1038/s41928-018-0059-3 ↗
6. Photonics for artificial intelligence and neuromorphic computing10.1038/s41566-020-00754-y ↗
7. Physics for neuromorphic computing10.1038/s42254-020-0208-2 ↗
8. The ‘echo state’ approach to analysing and training recurrent neural networks-with an erratum note
9. Information processing using a single dynamical node as complex system10.1038/ncomms1476 ↗
10. 2021
11. Constructive role of noise for high-quality replication of chaotic attractor dynamics using a hardware-based reservoir computer10.1103/physrevapplied.12.034058 ↗
12. High-speed photonic reservoir computing using a time-delay-based architecture: Million words per second classification10.1103/physrevx.7.011015 ↗
13. Chaos recognition using a single nonlinear node delay-based reservoir computer10.1140/epjb/s10051-022-00280-6 ↗
14. Classification of hyperchaotic, chaotic, and regular signals using single nonlinear node delay-based reservoir computers10.1063/5.0124204 ↗
15. A systematic literature review on prototyping with Arduino: Applications, challenges, advantages, and limitations10.1016/j.cosrev.2021.100364 ↗
16. An electronic analog of the Mackey–Glass system10.1016/0375-9601(95)00208-k ↗
17. Real-time audio processing with a cascade of discrete-time delay line-based reservoir computers10.1007/s12559-017-9457-5 ↗
18. Delay-based reservoir computing: Noise effects in a combined analog and digital implementation10.1109/tnnls.2014.2311855 ↗
19. Heart sound signal recovery based on time series signal prediction using a recurrent neural network in the long short-term memory model10.1007/s11227-019-03096-x ↗