In this paper, a scheme is proposed where the channel and spatial attention convolutional neural networks are applied to identify modulation formats from signal constellation diagrams. According to the simulation results, it outperforms other modulation format identification (MFI) schemes in the overall identification rate. Moreover, the scheme shows a significant advantage in signal identification given low optical signal-to-noise ratios (OSNRs).


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Modulation Format Identification Based on Channel-spatial Attention Modules and Deep Learning


    Beteiligte:
    Chen, Qian (Autor:in) / Zhang, Qi (Autor:in) / Xin, Xiangjun (Autor:in) / Cui, Yi (Autor:in) / Wang, Fu (Autor:in) / Tian, Feng (Autor:in) / Tian, Qinghua (Autor:in) / Wang, Yongjun (Autor:in) / Yang, Leijing (Autor:in)


    Erscheinungsdatum :

    2023-10-11


    Format / Umfang :

    3003205 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Signal power distribution based modulation format identification for coherent optical receivers

    Liu, Jie / Zhong, Kangping / Dong, Zhenhua et al. | British Library Online Contents | 2017


    Blind Density-Peak-Based Modulation Format Identification for Elastic Optical Networks

    Jiang, L. / Yan, L. / Yi, A. et al. | British Library Online Contents | 2018


    Blind Optical Modulation Format Identification From Physical Layer Characteristics

    Adles, E.J. / Dennis, M.L. / Johnson, W.R. et al. | British Library Online Contents | 2014