Learning to Equalize OTFS

被引:18
|
作者
Zhou, Zhou [1 ]
Liu, Lingjia [1 ]
Xu, Jiarui [1 ]
Calderbank, Robert [2 ]
机构
[1] Wireless VT, Bradley Dept Elect & Comp Engn, Blacksburg, VA 24061 USA
[2] Duke Univ, Dept Elect & Comp Engn, Durham, NC 27708 USA
关键词
OFDM; Training; Time-frequency analysis; Wireless communication; Channel estimation; Modulation; MIMO communication; OTFS; delay-Doppler; neural network; online learning; reservoir computing; one-shot learning; channel equalization; 5G-advanced; symbol detection; AIDED CHANNEL ESTIMATION; OFDM SYMBOL DETECTION; MIMO; SYSTEMS; PILOT;
D O I
10.1109/TWC.2022.3160600
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Orthogonal Time Frequency Space (OTFS) is a novel framework that processes modulation symbols via a time-independent channel characterized by the delay-Doppler domain. The conventional waveform, orthogonal frequency division multiplexing (OFDM), requires tracking frequency selective fading channels over the time, whereas OTFS benefits from full time-frequency diversity by leveraging appropriate equalization techniques. In this paper, we consider a neural network-based supervised learning framework for OTFS equalization. Learning of the introduced neural network is conducted in each OTFS frame fulfilling an online learning framework: the training and testing datasets are within the same OTFS-frame over the air. Utilizing reservoir computing, a special recurrent neural network, the resulting one-shot online learning is sufficiently flexible to cope with channel variations among different OTFS frames (e.g., due to the link/rank adaptation and user scheduling in cellular networks). The proposed method does not require explicit channel state information (CSI) and simulation results demonstrate a lower bit error rate (BER) than conventional equalization methods in the low signal-to-noise (SNR) regime under large Doppler spreads. When compared with its neural network-based counterparts for OFDM, the introduced approach for OTFS will lead to a better tradeoff between the processing complexity and the equalization performance.
引用
收藏
页码:7723 / 7736
页数:14
相关论文
共 50 条
  • [1] Should We Equalize Status in Order to Equalize Health?
    Nielsen, Morten Ebbe Juul
    Landes, Xavier
    Andersen, Martin Marchman
    PUBLIC HEALTH ETHICS, 2013, 6 (01) : 104 - 113
  • [2] Data-Driven Deep Learning for OTFS Detection
    Yi Gong
    Qingyu Li
    Fanke Meng
    Xinru Li
    Zhan Xu
    China Communications, 2023, 20 (01) : 88 - 101
  • [3] IS THERE A TENDENCY FOR PROFITS TO EQUALIZE?
    Tucker, Rufus S.
    AMERICAN ECONOMIC REVIEW, 1937, 27 (03): : 519 - 524
  • [4] Deep Learning-Based Downlink OTFS-SCMA
    Li, Qingyu
    Liu, Guanghui
    Zhao, Yanbo
    Liu, Chengxiang
    Xu, Fuchen
    IEEE WIRELESS COMMUNICATIONS LETTERS, 2024, 13 (12) : 3434 - 3438
  • [5] Data-Driven Deep Learning for OTFS Detection br
    Gong, Yi
    Li, Qingyu
    Meng, Fanke
    Li, Xinru
    Xu, Zhan
    CHINA COMMUNICATIONS, 2023, 20 (01) : 88 - 101
  • [6] CAPACITIES HARD TO EQUALIZE
    BIRKENBERTSCH, G
    WERKSTATTSTECHNIK ZEITSCHRIFT FUR INDUSTRIELLE FERTIGUNG, 1976, 66 (01): : 52 - 53
  • [7] FLOW EQUALIZE ALL INFLUENT
    UPDIKE, J
    WATER & WASTES ENGINEERING, 1980, 17 (10): : 20 - 23
  • [8] EQUALIZE PAD POWER DISSIPATIONS
    DUNN, J
    ELECTRONIC DESIGN, 1994, 42 (10) : 92 - &
  • [9] Equalize access to medical advances
    Dunne, Colum P.
    Dunne, Suzanne S.
    NATURE, 2018, 559 (7712) : 32 - 32
  • [10] SHAKING BEES TO EQUALIZE BROOD
    HAYES, B
    AMERICAN BEE JOURNAL, 1979, 119 (04): : 289 - 289