Presentation 2022-03-04
[Encouragement Talk] A Study on the Possibility of Estimating Multiple Communication Environment Information by Deep Learning
Shun Kojima, Kazuki Maruta, Yi Feng, Takashi Yokota, Kanemitsu Ootsu, Chang-Jun Ahn, Vahid Tarokh,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In the next generation mobile radio communication systems, it is essential to obtain the communication environment information accurately and quickly in order to implement appropriate control such as adaptive modulation and coding for realizing high-speed, high-capacity and low-delay communication. SNR, Doppler shift, and K-factor are some of the communication environment parameters that have a significant impact on the performance of adaptive modulation and coding. In the past, it has been difficult to introduce these parameters into adaptive modulation and coding for high-speed and large-capacity communications because the estimation of these parameters requires a huge amount of computation, a reference signal, and large-scale signal sampling. In this paper, we propose a method for estimating these multiple communication environment parameters on a per-packet basis without using reference signals by using convolutional neural networks from spectrogram images of the received signal. From the simulation results, we clarify the effectiveness of the proposed method in terms of the estimation accuracy of SNR, Doppler shift, and K-factor when they are estimated independently and the estimation accuracy when these three parameters are estimated simultaneously.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Spectrogram / CNN / SNR estimation / Doppler shift estimation / K-factor estimation
Paper # RCS2021-285
Date of Issue 2022-02-23 (RCS)

Conference Information
Committee RCS / SR / SRW
Conference Date 2022/3/2(3days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) Mobile Communication Workshop
Chair Eiji Okamoto(Nagoya Inst. of Tech.) / Suguru Kameda(Hiroshima Univ.) / Hanako Noda(Anritsu)
Vice Chair Toshihiko Nishimura(Hokkaido Univ.) / Tomoya Tandai(Toshiba) / Fumihide Kojima(NICT) / Osamu Takyu(Shinshu Univ.) / Kentaro Ishidu(NICT) / Kazuto Yano(ATR) / Keiichi Mizutani(Kyoto Univ.) / Kentaro Saito(Tokyo Denki Univ.) / Hirokazu Sawada(NICT)
Secretary Toshihiko Nishimura(NEC) / Tomoya Tandai(Panasonic) / Fumihide Kojima(Mie Univ.) / Osamu Takyu(Tokai Univ.) / Kentaro Ishidu(NTT) / Kazuto Yano(NTT) / Keiichi Mizutani(NIigata Univ.) / Kentaro Saito / Hirokazu Sawada
Assistant Koichi Adachi(Univ. of Electro-Comm.) / Osamu Nakamura(Sharp) / Manabu Sakai(Mitsubishi Electric) / Masashi Iwabuchi(NTT) / Tatsuki Okuyama(NTT DOCOMO) / Mai Ohta(Fukuoka Univ.) / Taichi Ohtsuji(NEC) / WANG Xiaoyan(Ibaraki Univ.) / Akemi Tanaka(MathWorks) / Akihito Noda(Nanzan Univ.)

Paper Information
Registration To Technical Committee on Radio Communication Systems / Technical Committee on Smart Radio / Technical Committee on Short Range Wireless Communications
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) [Encouragement Talk] A Study on the Possibility of Estimating Multiple Communication Environment Information by Deep Learning
Sub Title (in English)
Keyword(1) Spectrogram
Keyword(2) CNN
Keyword(3) SNR estimation
Keyword(4) Doppler shift estimation
Keyword(5) K-factor estimation
1st Author's Name Shun Kojima
1st Author's Affiliation Utsunomiya University(Utsunomiya Univ.)
2nd Author's Name Kazuki Maruta
2nd Author's Affiliation Tokyo Institute of Technology(Tokyo Tech.)
3rd Author's Name Yi Feng
3rd Author's Affiliation Aptiv(Aptiv)
4th Author's Name Takashi Yokota
4th Author's Affiliation Utsunomiya University(Utsunomiya Univ.)
5th Author's Name Kanemitsu Ootsu
5th Author's Affiliation Utsunomiya University(Utsunomiya Univ.)
6th Author's Name Chang-Jun Ahn
6th Author's Affiliation Chiba University(Chiba Univ.)
7th Author's Name Vahid Tarokh
7th Author's Affiliation Duke University(Duke Univ.)
Date 2022-03-04
Paper # RCS2021-285
Volume (vol) vol.121
Number (no) RCS-391
Page pp.pp.164-169(RCS),
#Pages 6
Date of Issue 2022-02-23 (RCS)