Presentation 2016-07-22
A Novel Modulation Classification Method in Cognitive Radios using Deep Network
Xu Zhu, Takeo Fujii,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a universal modulation classification method based on Denoise Stacked Sparse Auto-encoder (DSSA), one type of deep networks, which extracts features and classifies for single carrier modulation classification. This method can extract modulation features automatically, and classify input signals based on the features it extracted, which enables us to utilize it on as many modulations as we use in practice. Same as conventional neural network, a labeled samples base is necessary for parameters training. Network structure, however, is different for conventional neural network, since it can simplify an exponentially large number of hidden units by a multi-layer construction. This simplification enables us to achieve better back propagation and network tune. In addition, Denoising Auto-encoder extends the performance of Auto-encoder by reconstruct the data from a corrupted version to extract more robust feature. A series of results of binary classification, along with multi-classes classification are given by simulation, which shows an more universal utilization than other methods.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) modulation classificationcognitive radiodenoising autoencoder
Paper # SR2016-50
Date of Issue 2016-07-13 (SR)

Conference Information
Committee RCS / RCC / ASN / NS / SR
Conference Date 2016/7/20(3days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English) Wireless Distributed Network, M2M: Machine-to-Machine, D2D (Device-to-Device),etc.
Chair Hidekazu Murata(Kyoto Univ.) / Shinsuke Hara(Osaka City Univ.) / Hiroshi Tohjo(NTT) / Hideki Tode(Osaka Pref. Univ.) / Takeo Fujii(Univ. of Electro-Comm.)
Vice Chair Satoshi Denno(Okayama Univ.) / Yukitoshi Sanada(Keio Univ.) / Eisuke Fukuda(Fujitsu Labs.) / Kazunori Hayashi(Kyoto Univ.) / Ryu Miura(NICT) / Hiroo Sekiya(Chiba Univ.) / Hiraku Okada(Nagoya Univ.) / Satoru Yamano(NEC) / Yoshikatsu Okazaki(NTT) / Kenta Umebayashi(Tokyo Univ. of Agric. and Tech.) / Masayuki Ariyoshi(NEC) / Suguru Kameda(Tohoku Univ.)
Secretary Satoshi Denno(Toshiba) / Yukitoshi Sanada(NTT DoCoMo) / Eisuke Fukuda(Kagawa Univ.) / Kazunori Hayashi(Hokkaido Univ.) / Ryu Miura(Kanagawa Inst. of Tech.) / Hiroo Sekiya(NTT) / Hiraku Okada(Kyushu Inst. of Tech.) / Satoru Yamano(NTT) / Yoshikatsu Okazaki(Shinshu Univ.) / Kenta Umebayashi(NICT) / Masayuki Ariyoshi / Suguru Kameda
Assistant Tetsuya Yamamoto(Panasonic) / Toshihiko Nishimura(Hokkaido Univ.) / Koichi Ishihara(NTT) / Kazushi Muraoka(NEC) / Shinsuke Ibi(Osaka Univ.) / Toshinori Kagawa(NICT) / Kentaro Kobayashi(Nagoya Univ.) / Yuichi Igarashi(Hitachi) / Katsuhiro Naito(Aichi Inst. of Tech.) / Kiyohiko Hattori(NICT) / Hiroshi Fujita(Fujitsu Labs.) / Takuro Yonezawa(Keio Univ.) / Shohei Kamamura(NTT) / Kazuto Yano(ATR) / Mamiko Inamori(Tokai Univ.) / Hiroyuki Shiba(NTT) / Gia Khanh Tran(Tokyo Inst. of Tech.)

Paper Information
Registration To Technical Committee on Radio Communication Systems / Technical Committee on Reliable Communication and Control / Technical Committee on Ambient intelligence and Sensor Networks / Technical Committee on Network Systems / Technical Committee on Smart Radio
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Novel Modulation Classification Method in Cognitive Radios using Deep Network
Sub Title (in English)
Keyword(1) modulation classificationcognitive radiodenoising autoencoder
1st Author's Name Xu Zhu
1st Author's Affiliation The University of Electro-Communications(UEC)
2nd Author's Name Takeo Fujii
2nd Author's Affiliation The University of Electro-Communications(UEC)
Date 2016-07-22
Paper # SR2016-50
Volume (vol) vol.116
Number (no) SR-148
Page pp.pp.103-106(SR),
#Pages 4
Date of Issue 2016-07-13 (SR)