Paper Abstract and Keywords |
Presentation |
2019-07-11 09:35
A study on Auto-Regressive modeling of Duty Cycle Kohei Okawa, Hiroki Iwata, Kenta Umebayashi (Tokyo Univ. of Agriculture and Tech.), Janne Lehtomäki (Univ. of Oulu), Miguel López-Benítez (Univ. of Liver), Satya Joshi (Univ. of Oulu) SR2019-29 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
In dynamic spectrum sharing, it is useful to exploit statistical information on spectrum usage.
In this paper, we investigate an AR (Auto-Regressive) modeling to predict DC (Duty Cycle) which is spectrum occupancy rate over time axis.
Real DC measurements show behavior that is hard for the typical AR model to express such as a time-varying characteristic of mean DC and existence of time period with low-correlation.
To overcome the behavior, we propose a DC model based on AR model considering these behavior.
Specifically, the proposed model classifies the states of DC value (high or low) exploiting thresholding process and models the DC based on AR model each state.
We also consider the time-varying characteristic of mean DC by means of estimation of mean DC by sliding window process.
Several verifications based on real DC measurements show the proposed model can attain high enough prediction accuracy in case of long enough observation period of DC. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
Cognitive radio / dynamic spectrum access / smart spectrum access / duty cycle / auto-regressive model / / / |
Reference Info. |
IEICE Tech. Rep., vol. 119, no. 109, SR2019-29, pp. 59-64, July 2019. |
Paper # |
SR2019-29 |
Date of Issue |
2019-07-03 (SR) |
ISSN |
Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
SR2019-29 |
|