Electronic System for Chaotic Time Series Prediction Associated to Human Disease

It is well-known that a large number of natural phenomena exhibit chaotic behavior, e.g. brain activity, mental illness, bioelectric signals, pancreatic beta cell, and so on. That way, researchers have the challenge to develop systems that guarantee the prediction of chaotic time series, so that it...

Full description

Saved in:
Bibliographic Details
Main Authors: Pano-Azucena, Ana Dalia, Tlelo-Cuautle, Esteban, Tan, Sheldon
Format: Conference Proceeding
Language:English
Subjects:
Chaotic oscillator
Field programmable gate arrays
FPGA
Hardware
multilayer perceptron
Neurons
Oscillators
Time series analysis
time series prediction
Topology
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:It is well-known that a large number of natural phenomena exhibit chaotic behavior, e.g. brain activity, mental illness, bioelectric signals, pancreatic beta cell, and so on. That way, researchers have the challenge to develop systems that guarantee the prediction of chaotic time series, so that it can be used to prevent the activation of an epileptic attack or other human disorder. In this paper we show the usefulness of the multilayer perceptron (MLP), which is in the family of artificial neural networks, to predict chaotic time series, which in this research paper were obtained from real chaotic systems based on saturated nonlinear function series and from the Rossler system. We highlight the hardware implementation of the prediction system that is verified by using a field-programmable gate array (FPGA). The root-mean-square error is provided to show the suitability of the proposed electronic system.
ISSN:2575-2634
DOI:10.1109/ICHI.2018.00044