VANET–LTE based heterogeneous vehicular clustering for driving assistance and route planning applications

Iftikhar Ahmad, Rafidah Md Noor, Ismail Ahmedy, Syed Adeel Ali Shah, Ibrar Yaqoob, Ejaz Ahmed, Muhammad Imran

Research output: Contribution to journalArticlepeer-review

33 Scopus citations

Abstract

The Internet of vehicles incorporates multiple access networks and technologies to connect vehicles on roads. These vehicles usually require the use of individual long-term evolution (LTE) connections to send/receive data to/from a remote server to make smart decisions regarding route planning and driving. An increasing number of vehicles on the roads may not only overwhelm LTE network usage but also incur added cost. Clustering helps minimize LTE usage, but the high speed of vehicles renders connections unstable and unreliable not only among vehicles but also between vehicles and the LTE network. Moreover, non-cooperative behavior among vehicles within a cluster is a bottleneck in sharing costly data acquired from the Internet. To address these issues, we propose a novel destination- and interest-aware clustering (DIAC) mechanism. DIAC primarily incorporates a strategic game-theoretic algorithm and a self-location calculation algorithm. The former allows vehicles to participate/cooperate and enforces a fair-use policy among the cluster members (CMs), whereas the latter enables CMs to calculate their location coordinates in the absence of a global positioning system under an urban topography. DIAC strives to reduce the frequency of link failures not only among vehicles but also between each vehicle and the 3G/LTE network. The mechanism also considers vehicle mobility and LTE link quality and exploits common interests among vehicles in the cluster formation phase. The performance of the DIAC mechanism is validated through extensive simulations, whose results demonstrate that the performance of the proposed mechanism is superior to that of similar and existing approaches.

Original languageBritish English
Pages (from-to)128-140
Number of pages13
JournalComputer Networks
Volume145
DOIs
StatePublished - 9 Nov 2018

Keywords

  • Clustering
  • Cooperative communication
  • Internet of vehicles
  • Long-term evolution
  • Vehicular ad hoc networks

Fingerprint

Dive into the research topics of 'VANET–LTE based heterogeneous vehicular clustering for driving assistance and route planning applications'. Together they form a unique fingerprint.

Cite this