Bloom Hopping: Bloom Filter Based 2-Hop Neighbor Management in VANETs

Florian Klingler, Reuven Cohen, Christoph Sommer, Falko Dressler

Research output: Contribution to journalArticlepeer-review

Abstract

Recent works have shown that it would be beneficial for nodes in wireless networks with very dynamic topology to maintain a list of 2-hop neighbors, namely, the neighbors of its neighbors. This is important, for example, for routing, clustering, and message dissemination to all the nodes in a given geographic vicinity. In this paper, we propose a scheme that uses Bloom filters for maintaining 2-hop neighborship information. Furthermore, we developed a novel 2-hop broadcast algorithm making use of the specific nature of our Bloom filter encoded neighbor information. We particularly focus on the Vehicular Ad Hoc Networks (VANETs) application scenario. Here, beaconing is a periodic broadcast of awareness messages by each vehicle to its immediate neighbors. A naïve approach would be to include all 2-hop neighbors in each beacon message, which, however, would work only for small or sparse scenarios. We show that our approach significantly reduces the length of the beacon messages, thereby keeping channel load and collision probability considerably lower than in the naïve scheme. We further demonstrate the application of our Bloom filter based 2-hop neighbor table for developing higher layer protocols and introduce a multi-hop broadcast protocol called Bloom Hopping.

Original languageEnglish
Article number8364590
Pages (from-to)534-545
Number of pages12
JournalIEEE Transactions on Mobile Computing
Volume18
Issue number3
DOIs
StatePublished - 1 Mar 2019

Keywords

  • Vehicular networking
  • bloom filter
  • broadcast
  • neighbor management
  • vehicular ad hoc network

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Bloom Hopping: Bloom Filter Based 2-Hop Neighbor Management in VANETs'. Together they form a unique fingerprint.

Cite this