A Waveform Parameter Assignment Framework for 6G With the Role of Machine Learning

Creative Commons License

YAZAR A., Arslan H.

IEEE OPEN JOURNAL OF VEHICULAR TECHNOLOGY, cilt.1, ss.156-172, 2020 (ESCI) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 1
  • Basım Tarihi: 2020
  • Doi Numarası: 10.1109/ojvt.2020.2992502
  • Dergi Adı: IEEE OPEN JOURNAL OF VEHICULAR TECHNOLOGY
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus
  • Sayfa Sayıları: ss.156-172
  • Anahtar Kelimeler: 6G, beyond 5G, machine learning, multiple numerologies, OFDM, radio resource management, scheduling, waveform, WIRELESS NETWORKS, CHANNEL ESTIMATION, BIG DATA, SIGNAL IDENTIFICATION, RESOURCE-ALLOCATION, MIXED NUMEROLOGIES, INDEX MODULATION, NEXT-GENERATION, MULTIPLE-ACCESS, NEURAL-NETWORKS
  • Eskişehir Osmangazi Üniversitesi Adresli: Hayır

Özet

5G enables a wide variety of wireless communications applications and use cases. There are different requirements associated with the applications, use cases, channel structure, network and user. To meet all of the requirements, several new configurable parameters are defined in 5G New Radio (NR). It is possible that 6G will have even higher number of configurable parameters based on new potential conditions. In line with this trend, configurable waveform parameters are also varied and this variation will increase in 6G considering the potential future necessities. In this paper, association of users and possible configurable waveform parameters in a cell is discussed for 6G communication systems. An assignment framework of configurable waveform parameters with different types of resource allocation optimization mechanisms is proposed. Most of all, the role and usage of machine learning (ML) in this framework is described. A case study with a simulation based dataset generation methodology is also presented.