Versatile prediction core for IoT applications

A novel prediction method which can be applied for wide range of IoT applications is presented. Conventional prediction methods have problems such as that they are not suitable for long-term prediction, and that processing time increases as the number of tracking target becomes large. The proposed m...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE International Conference on Consumer Electronics (ICCE) pp. 1 - 2
Main Authors Uejima, Toshiaki, Mizosoe, Hiroki, Suzuki, Kunihiko, Yamada, Takashi
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.01.2018
Subjects
Automobiles
Dictionaries
Machine learning
Particle filters
Prediction methods
Sensors
Target tracking
Online AccessGet full text

Cover

More Information
Summary:A novel prediction method which can be applied for wide range of IoT applications is presented. Conventional prediction methods have problems such as that they are not suitable for long-term prediction, and that processing time increases as the number of tracking target becomes large. The proposed method estimates both the state and the state change of the tracked targets by matching multidimensional feature information gathered from a number of external sensors with internal dictionaries, which are built using machine learning in advance. The experimental results show that the proposed method achieves longer-term prediction with less computational time than a conventional method.
ISSN:2158-4001
DOI:10.1109/ICCE.2018.8326227