Presentation 2018-01-19
Machine Learning-based Primary Exclusive Region Update for Spectrum Sharing
Aogu Yamada, Takayuki Nishio, Masahiro Morikura, Koji Yamamoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In spectrum sharing, secondary users (SUs) utilize a licensed frequency band under a condition to avoid interference with a primary user (PU), such as a primary exclusive region (PER), wherein the SUs are prohibited to transmit. However, when SUs outside the PER interfere with the PU due to long term variation of propagation paths, the PER needs to be updated. In this paper, we propose a framework for updating the PER adaptively with supervised learning, when interference occurs. This frameworks employs training data sets, which are transmission logs of SUs with supervised labels whether or not PU detects interference from SUs. However, dominance of the non-interference labels compared to the interference labels degrades learning performance. To overcome imbalance of the supervised labels, we designed a novel sampling schemes considering propagation characteristics. Simulation results show the proposed framework continues to reduce interference probability and achieves smaller PER with fewer iterations compared to the conventional frameworks.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Machine learning / Spectrum sharing / Cognitive radio / Supervised learning / Imbalanced data
Paper # MoNA2017-52
Date of Issue 2018-01-11 (MoNA)

Conference Information
Committee MoNA
Conference Date 2018/1/18(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Campus Plaza Kyoto
Topics (in Japanese) (See Japanese page)
Topics (in English) Mobile Network, Application of Machine Learning, Mobile Data, etc.
Chair Ryoichi Shinkuma(Kyoto Univ.)
Vice Chair Shigeaki Tagashira(Kansai Univ.) / Gen Kitagata(Tohoku Univ.)
Secretary Shigeaki Tagashira(Kyushu Univ.) / Gen Kitagata(NTT)
Assistant Takayuki Nishio(Kyoto Univ.) / Takato Saito(NTT)

Paper Information
Registration To Technical Committee on Mobile Network and Applications
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Machine Learning-based Primary Exclusive Region Update for Spectrum Sharing
Sub Title (in English)
Keyword(1) Machine learning
Keyword(2) Spectrum sharing
Keyword(3) Cognitive radio
Keyword(4) Supervised learning
Keyword(5) Imbalanced data
1st Author's Name Aogu Yamada
1st Author's Affiliation Kyoto University(Kyoto Univ.)
2nd Author's Name Takayuki Nishio
2nd Author's Affiliation Kyoto University(Kyoto Univ.)
3rd Author's Name Masahiro Morikura
3rd Author's Affiliation Kyoto University(Kyoto Univ.)
4th Author's Name Koji Yamamoto
4th Author's Affiliation Kyoto University(Kyoto Univ.)
Date 2018-01-19
Paper # MoNA2017-52
Volume (vol) vol.117
Number (no) MoNA-390
Page pp.pp.63-68(MoNA),
#Pages 6
Date of Issue 2018-01-11 (MoNA)