Fault-aware flow control and multi-path routing in VANETs

Xiaomei Zhang, Xiaolei Dong, Naixue Xiong, Jie Wu, Xiuqi Li

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

Built upon the fragile mobile devices and shared wireless channels, Vehicular Ad-Hoc Networks (VANETs) are vulnerable to misbehaving nodes. The critical problems that affect the serviceability of the entire network are maintaining an acceptable level of network performance degradation, and fair allocation of resource among different users in the presence of misbehaving nodes. This paper investigates the optimization problem via the joint design of rate control and multi-path routing in fault-aware VANETs. We use statistical information and estimation on each wireless link to characterize the effect of faults, and develop a leaky-path model. This model takes account of packet loss along fault-paths and the “shrinking” feature of effective flow at the destination node. Moreover, we use a cost function to measure the impact of fault-correlation among multi-routing paths on effective flows. Based on the leaky-path model and the cost function, a fault-correlated flow control and routing ((FC)2R) approach is proposed to maximize the network utility associated with the effective rate. We design a novel distributed algorithm to adjust flow rates adaptively on each path, using multi-path routing. Our simulation results demonstrate that higher effective network throughput, and better fairness, can be achieved by our algorithm than the standard optimal flow control in the presence of misbehaving nodes.

Original languageEnglish
Pages (from-to)1090-1107
Number of pages18
JournalPeer-to-Peer Networking and Applications
Volume8
Issue number6
DOIs
StatePublished - 21 Nov 2015
Externally publishedYes

Keywords

  • Misbehaving node
  • Rate control
  • Routing
  • Vehicular Ad-Hoc networks

Fingerprint

Dive into the research topics of 'Fault-aware flow control and multi-path routing in VANETs'. Together they form a unique fingerprint.

Cite this