SoNaR: Software-Defined Network and Radio Framework for FANETs


Çoğay S., SARI T. T., SEÇİNTİ G.

IEEE International Symposium on Dynamic Spectrum Access Networks, Amerika Birleşik Devletleri, 13 - 15 Aralık 2021 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/dyspan53946.2021.9677087
  • Basıldığı Ülke: Amerika Birleşik Devletleri
  • Anahtar Kelimeler: Ad-hoc networks, proactive routing, dynamic spectrum access
  • İstanbul Teknik Üniversitesi Adresli: Evet

Özet

In communication networks, where the same channel is utilized for both control and data traffic, proactive routing protocols may lead high latency and packet loss values even when network size slightly increases. This drawback may easily become intolerable, especially in the networks with high mobility features due to the necessity of frequent control message broadcast in order to keep the routing tables up-to-date. As one of the prominent and unique examples of mobile networks of the last decade, flying ad-hoc networks (FANETs) also suffer maintaining e2e communication paths using the existing routing and access protocols. In this paper, we propose a framework, namely SoNaR, which leverages flexibility of software-defined approaches both in network and channel access layers. In this manner, we employ SDN architecture, where ground control station of the aerial nodes/drones acting also as a network controller and utilize on-board SDRs allowing control and data traffic to co-exist on the same spectrum with different modulation techniques. Our framework allows the drones to use on-board SDRs managing direct-sequence spread spectrum (DSSS) links and to use COTS dongles supports IEEE 802.11 IBSS using OFDM for control and data messages respectively, where both transmitted in 2.4 GHz band. Furthermore, we propose an hybrid routing and channel selection, based on branch and bound algorithm, which solves routing and channel allocation challenges in conjunction harnessing the flexibility offered by our framework. Our simulation results show that, our method offers 51.22% increase in throughput with 52.44% less latency compared to a base case that only uses IEEE 802.11 IBSS links for both data and control traffic.