On the Investigation of Bandwidth Allocation for Short-Range and Long-Range Heterogeneous Wireless Networks


Kazdaǧ C., Alaca Ö., Ekti A. R., YARKAN S.

16th International Multi-Conference on Systems, Signals and Devices, SSD 2019, İstanbul, Türkiye, 21 - 24 Mart 2019, ss.524-528 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/ssd.2019.8893229
  • Basıldığı Şehir: İstanbul
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.524-528
  • Anahtar Kelimeler: HetNets, MT assignment, bandwidth allocation, multi-homing, MINLP
  • İstanbul Ticaret Üniversitesi Adresli: Evet

Özet

It is clear that heterogeneous wireless network (HetNet) will be an essential deployment mode of next generation wireless networks. There are various HetNet layouts present in the literature such as coexistences of mid-range networks and long-range networks. With the recent advances in digital technology, mobile terminals (MTs), which are the integral part of HetNet, have multiple radio interfaces which brings the concept of multi-homing capabilities. As a result, MTs are able to establish connections to multiple wireless networks. MTs also can increase the bandwidth by benefiting from all these available networks around to meet the requirements of demanding applications such as high definition multimedia stream. Such a diverse set of scenarios brings about a critical challenge regarding how MT assignment and bandwidth allocation could be carried out jointly in HetNet. In contrast to the traditional layouts present in the literature, in this study, a more realistic scenario is investigated by considering the coexistence of mid- and long-range networks with an extension of the presence of short-range networks. Therefore, as a solution mixed integer non-linear program (MINLP) approach is proposed by incorporating the connection interfaces of Bluetooth, wireless-fidelity (WiFi) and cellular systems. It is shown with both the aid of mathematical analysis and the support of numerical results that the method proposed attains suboptimal MT assignment and bandwidth allocation jointly with significantly reduced computational complexity.