A Distributed Mobile Fog Computing Scheme for Mobile Delay-Sensitive Applications in SDN-Enabled Vehicular Networks

With the rapid development of intelligent transportation systems, enormous amounts of delay-sensitive vehicular services have been emerging and challenge both the architectures and protocols of vehicular networks. However, existing cloud computing-embedded vehicular networks cannot guarantee timely data processing or service access, due to long propagation delay and traffic congestion at the cloud center. Meanwhile, the current distributed network architecture does not support scalable network management, leading the intelligent data computing policies to be undeployable. With this motivation, the authors propose to introduce fog computing into vehicular networks and define the Multiple Time-constrained Vehicular applications Scheduling (MTVS) issue. First, to improve the network flexibility and controllability, the authors introduce a Fog-based Base Station (FBS) and propose a Software-Defined Networking (SDN)-enabled architecture dividing the networks into network, fog, and control layers. To address MTVS issue, instead of normal centralized computing-based approaches, the authors propose to distribute mobile delay-sensitive task in data-level over multiple FBSs. In particular, the authors regard the fog layer of SDN-enabled network as an FBS-based network and propose to distribute the computing task based on the FBSs along multiple paths in the fog layer. By Linear Programming, the authors optimize the optimal data distribution/transmission model by formulating the delay computation model. Then, the authors propose a hybrid scheduling algorithm including both local scheduling and fog scheduling, which can be deployed on the proposed SDN-enabled vehicular networks. Simulation results demonstrate that the authors' approach performs better than some recent research outcomes, especially in the success rate for addressing MTVS issue.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01748248
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Aug 7 2020 2:10PM