• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Data-driven Algorithm Design for Wireless Communications

Research Project

Project/Area Number 19H02138
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 21020:Communication and network engineering-related
Research InstitutionNagoya Institute of Technology

Principal Investigator

Wadayama Tadashi  名古屋工業大学, 工学(系)研究科(研究院), 教授 (20275374)

Co-Investigator(Kenkyū-buntansha) 林 和則  京都大学, 国際高等教育院, 教授 (50346102)
Project Period (FY) 2019-04-01 – 2022-03-31
Project Status Completed (Fiscal Year 2021)
Budget Amount *help
¥12,740,000 (Direct Cost: ¥9,800,000、Indirect Cost: ¥2,940,000)
Fiscal Year 2021: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2020: ¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000)
Keywords深層学習 / 機械学習 / 無線通信 / 信号処理 / 深層展開 / 無線信号処理 / MIMO検出 / LDPC符号 / 圧縮センシング / SCDMA方式 / 収束加速 / MIMO方式 / 固定点反復 / チェビシェフ多項式 / スパース信号再現 / MIMO信号処理 / スパースCDMA
Outline of Research at the Start

本研究の目的は、無線通信系アルゴリズムに対するデータ駆動アルゴリズムデザインの原理を確立し、その原理に基づいて従前のアルゴリズムより優れた特性を持つアルゴリズムを創出することである。最適推定原理から演繹的に導出された反復推定アルゴリズムに対して、適切に学習可能パラメータを組み込み、さらに訓練データに基づいてそれらのパラメータを調整する手法をデータ駆動アルゴリズムデザインと呼ぶ。本研究では、様々な角度からデータ駆動アルゴリズムデザインの可能性を追求する。

Outline of Final Research Achievements

A data-driven approach for developing signal processing algorithms for wireless communications has been the primal target of this project. We have established the methodology called ``deep unfolding'' for constructing a new signal processing algorithms. The outputs of this project can be classified into two categories: 1) applications of deep unfolding for signal processing algorithms, 2) theoretical studies regarding the trained results. The first category includes our sparse signal reconstruction algorithm TISTA and trainable signal detection algorithms for MIMO wireless communications. The second category contains the works regarding Chebyshev steps, which can explains the trained results of deep unfolded gradient descent.

Academic Significance and Societal Importance of the Research Achievements

本研究の目的は、無線通信系アルゴリズムに対するデータ駆動アルゴリズムデザインの原理を確立し、その原理に基づいて従前のアルゴリズムより優れた特性を持つアルゴリズムを創出することである。最適推定原理から演繹的に導出された反復推定アルゴリズムに対して、適切に学習可能パラメータを組み込み、さらに訓練データに基づいてそれらのパラメータを調整する手法をデータ駆動アルゴリズムデザインと呼ぶ。通常の反復推定アルゴリズムを展開表現することで多層ニューラルネットワークと見立て、与えられた訓練データに基づき深層学習に基づくパラメータの調整を行うことにより、従来法では達成が困難であった性能が実現できる。

Report

(4 results)
  • 2021 Annual Research Report   Final Research Report ( PDF )
  • 2020 Annual Research Report
  • 2019 Annual Research Report
  • Research Products

    (20 results)

All 2022 2021 2020 2019

All Journal Article (10 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 10 results,  Open Access: 6 results) Presentation (10 results) (of which Int'l Joint Research: 10 results)

  • [Journal Article] Convergence Acceleration via Chebyshev Step: Plausible Interpretation of Deep-Unfolded Gradient Descent2022

    • Author(s)
      S. Takabe and T. Wadayama
    • Journal Title

      IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

      Volume: to appear

    • NAID

      130008144000

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Chebyshev Periodical Successive Over-Relaxation for Accelerating Fixed-Point Iterations2021

    • Author(s)
      T. Wadayama and S. Takabe
    • Journal Title

      IEEE Signal Processing Letters

      Volume: 28

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Transmitter and receiver impairment monitoring using adaptive multi-layer linear and widely linear filter coefficients controlled by stochastic gradient descent2021

    • Author(s)
      M. Arikawa and K. Hayashi
    • Journal Title

      Optics Express

      Volume: 29 Issue: 8 Pages: 11548-11561

    • DOI

      10.1364/oe.416992

    • Related Report
      2021 Annual Research Report 2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Adaptive multi-layer filters incorporated with Volterra filters for impairment compensation including transmitter and receiver nonlinearity2021

    • Author(s)
      M. Arikawa and K. Hayashi
    • Journal Title

      Optics Express

      Volume: 29 Issue: 18 Pages: 28366-28387

    • DOI

      10.1364/oe.435161

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Deep-Unfolded Sparse CDMA: Multiuser Detector and Sparse Signature Design2021

    • Author(s)
      Takabe Satoshi、Yamauchi Yuki、Wadayama Tadashi
    • Journal Title

      IEEE Access

      Volume: 9 Pages: 40027-40038

    • DOI

      10.1109/access.2021.3064558

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Asymptotic Behavior of Spatial Coupling LDPC Coding for Compute-and-Forward Two-Way Relaying2020

    • Author(s)
      S. Takabe, and T. Wadayama, and M. Hayashi
    • Journal Title

      IEEE Transactions on Communications

      Volume: - Issue: 7 Pages: 4063-4072

    • DOI

      10.1109/tcomm.2020.2987891

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Deep Learning-Based Average Consensus2020

    • Author(s)
      Kishida Masako、Ogura Masaki、Yoshida Yuichi、Wadayama Tadashi
    • Journal Title

      IEEE Access

      Volume: 8 Pages: 142404-142412

    • DOI

      10.1109/access.2020.3014148

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Adaptive equalization of transmitter and receiver IQ skew by multi-layer linear and widely linear filters with deep unfolding2020

    • Author(s)
      Arikawa Manabu、Hayashi Kazunori
    • Journal Title

      Optics Express

      Volume: 28 Issue: 16 Pages: 23478-23478

    • DOI

      10.1364/oe.395361

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Trainable Projected Gradient Detector for Massive Overloaded MIMO Channels: Data-Driven Tuning Approach2019

    • Author(s)
      Satoshi Takabe, Masayuki Imanishi, Tadashi Wadayama, Ryo Hayakawa, Kazunori Hayashi
    • Journal Title

      IEEE Access

      Volume: 7 Pages: 93326-93338

    • DOI

      10.1109/access.2019.2927997

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Trainable ISTA for Sparse Signal Recovery2019

    • Author(s)
      D. Ito, S. Takabe, and T. Wadayama
    • Journal Title

      IEEE Transactions on Signal Processing

      Volume: 67 Issue: 12 Pages: 3113-3125

    • DOI

      10.1109/tsp.2019.2912879

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Presentation] Refined Density Evolution Analysis of LDPC Codes for Successive Interference Cancellation2021

    • Author(s)
      Satoshi Takabe and Tadashi Wadayama
    • Organizer
      IEEE Globecom
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] MSE-Optimaized Linear Transform for Noisy Fronthaul Channels in Distributed MIMO C-RAN2021

    • Author(s)
      Tadashi Wadayama and Satoshi Takabe
    • Organizer
      IEEE Globecom
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Proximal Decoding for LDPC-coded Massive MIMO Channels2021

    • Author(s)
      Tadashi Wadayama and Satoshi Takabe,
    • Organizer
      IEEE International Symposiumn on Information Theory
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] An Overloaded MU-MIMO Signal Detection Method Using Piecewise Continuous Nonconvex Sparse Regularizer2021

    • Author(s)
      A. Hirayama and K. Hayashi
    • Organizer
      APSIPA Annual Summit and Conference (APSIPA ASC 2021)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Deep Unfolded Multicast Beamforming2020

    • Author(s)
      S. Takabe and T. Wadayama
    • Organizer
      IEEE Globecom 2020
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Trainable Projected Gradient Detector for Sparsely Spread Code Division Multiple Access2020

    • Author(s)
      S. Takabe, Y. Yamauchi and T. Wadayama
    • Organizer
      IEEE International Conference on Communications (ICC 2020)
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Complex Trainable ISTA for Linear and Nonlinear Inverse Problems,2020

    • Author(s)
      S. Takabe, T. Wadayama, and Y. C. Eldar
    • Organizer
      IEEE ICASSP 2020
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Deep Learning-Aided Trainable Projected Gradient Decoding for LDPC Codes2019

    • Author(s)
      T. Wadayama and S. Takabe
    • Organizer
      IEEE International Symposium on Information Theory
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Data-Driven Tuning of Projected Gradient Algorithms for Signal Recovery Problems2019

    • Author(s)
      T. Wadayama
    • Organizer
      11th Asia-Europe Workshop on Concepts in Information Theory
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Deep Learning-Aided Projected Gradient Detector for Massive Overloaded MIMO Channels2019

    • Author(s)
      S. Takabe, M. Imanishi, T. Wadayama and K. Hayashi,
    • Organizer
      IEEE International Conference on Communications
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research

URL: 

Published: 2019-04-18   Modified: 2023-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi