Effects of diagram plane on neural network based modulation recognition

Basit Öğe Kaydı
dc.authoridÇalhan, Ali/0000-0002-5798-3103en_US
dc.authoridLeblebici, Merih/0000-0002-7709-2906en_US
dc.authoridCİCİOĞLU, MURTAZA/0000-0002-5657-7402en_US
dc.authorscopusid58723537700en_US
dc.authorscopusid16548877100en_US
dc.authorscopusid57203170833en_US
dc.authorwosidÇalhan, Ali/H-1375-2014en_US
dc.authorwosidLeblebici, Merih/HZK-6228-2023en_US
dc.authorwosidCİCİOĞLU, MURTAZA/AAL-5004-2020en_US
dc.contributor.authorLeblebici, Merih
dc.contributor.authorCalhan, Ali
dc.contributor.authorCicioglu, Murtaza
dc.date.accessioned2024-08-23T16:04:56Z
dc.date.available2024-08-23T16:04:56Z
dc.date.issued2024en_US
dc.departmentDüzce Üniversitesien_US
dc.description.abstractModulation recognition using deep learning presents challenges in effectively distinguishing high -order modulation schemes while maintaining a balance between complexity and recognition accuracy. In this study, we curate a comprehensive dataset in the r theta plane, encompassing eight distinct modulation schemes. Leveraging hyperparameter optimization and transfer learning, we explore the capabilities of various CNN -based architectures, including MobileNetV2, ResNet50V2, ResNet101V2, InceptionV3, ResNet152V2, Xception, and InceptionResNetV2, for the classification of modulation schemes. The simulation results demonstrate that with signalto-noise ratio (SNR) values exceeding 5 dB, all models exhibit classification accuracies surpassing 50% and approach near -perfect accuracy at an SNR value of 20 dB. However, under low SNR conditions, such as 5 dB, the recognition accuracies of all models, except for ResNet152V2 and InceptionV3, show minimal variation. As the SNR increases by 5 dB from -5 dB to 20 dB, ResNet152V2 and InceptionV3 demonstrate remarkable classification accuracy improvements, exceeding 40%, 30%, 30%, 10%, and 15%, respectively. In contrast, the other models do not exhibit such robust responsiveness in accuracy enhancements. The remarkable performance improvements are achieved by fine-tuning pre -trained models through these processes.en_US
dc.description.sponsorshipDzce University Scientific Research Project [BAP-2023.06.01.1410]en_US
dc.description.sponsorshipAcknowledgments This work was supported by the Duzce University Scientific Research Project under Grant BAP-2023.06.01.1410.en_US
dc.identifier.doi10.1016/j.asoc.2024.111412
dc.identifier.issn1568-4946
dc.identifier.issn1872-9681
dc.identifier.scopus2-s2.0-85185781736en_US
dc.identifier.scopusqualityQ1en_US
dc.identifier.urihttps://doi.org/10.1016/j.asoc.2024.111412
dc.identifier.urihttps://hdl.handle.net/20.500.12684/14415
dc.identifier.volume154en_US
dc.identifier.wosWOS:001190913100001en_US
dc.identifier.wosqualityN/Aen_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherElsevieren_US
dc.relation.ispartofApplied Soft Computingen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectModulation recognitionen_US
dc.subjectDeep learningen_US
dc.subjectTransfer learningen_US
dc.subjectConstellation diagramen_US
dc.subjectClassificationen_US
dc.titleEffects of diagram plane on neural network based modulation recognitionen_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu