Stochastic modeling for intelligent software defined vehicular networks: a survey

Banoth Ravi, Blesson Varghese, Ilir Murturi*, Praveen Kumar Donta*, Schahram Dustdar, Chinmaya Kumar Dehury, Satish Narayana Sriram*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)
11 Downloads (Pure)

Abstract

Digital twins and the Internet of Things (IoT) have gained significant research attention in recent years due to their potential advantages in various domains, and vehicular ad hoc networks (VANETs) are one such application. VANETs can provide a wide range of services for passengers and drivers, including safety, convenience, and information. The dynamic nature of these environments poses several challenges, including intermittent connectivity, quality of service (QoS), and heterogeneous applications. Combining intelligent technologies and software-defined networking (SDN) with VANETs (termed intelligent software-defined vehicular networks (iSDVNs)) meets these challenges. In this context, several types of research have been published, and we summarize their benefits and limitations. We also aim to survey stochastic modeling and performance analysis for iSDVNs and the uses of machine-learning algorithms through digital twin networks (DTNs), which are also part of iSDVNs. We first present a taxonomy of SDVN architectures based on their modes of operation. Next, we survey and classify the state-of-the-art iSDVN routing protocols, stochastic computations, and resource allocations. The evolution of SDN causes its complexity to increase, posing a significant challenge to efficient network management. Digital twins offer a promising solution to address these challenges. This paper explores the relationship between digital twins and SDN and also proposes a novel approach to improve network management in SDN environments by increasing digital twin capabilities. We analyze the pitfalls of these state-of-the-art iSDVN protocols and compare them using tables. Finally, we summarize several challenges faced by current iSDVNs and possible future directions to make iSDVNs autonomous.
Original languageEnglish
Article number162
Number of pages34
JournalComputers
Volume12
Issue number8
DOIs
Publication statusPublished - 12 Aug 2023

Keywords

  • Internet of Things
  • Vehicular ad hoc networks
  • Software-defined networks
  • Intelligent digital twin networks
  • Stochastic modeling and performance evaluation

Fingerprint

Dive into the research topics of 'Stochastic modeling for intelligent software defined vehicular networks: a survey'. Together they form a unique fingerprint.

Cite this