eduzhai > Applied Sciences > Engineering >

Energy-aware Graph Job Allocation in Software Defined Air-Ground Integrated Vehicular Networks

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 14 pages

Abstract: The software defined air-ground integrated vehicular (SD-AGV) networks haveemerged as a promising paradigm, which realize the flexible on-ground resourcesharing to support innovative applications for UAVs with heavy computationaloverhead. In this paper, we investigate a vehicular cloud-assisted graph joballocation problem in SD-AGV networks, where the computation-intensive jobscarried by UAVs, and the vehicular cloud are modeled as graphs. To map eachcomponent of the graph jobs to a feasible vehicle, while achieving thetrade-off among minimizing UAVs job completion time, energy consumption, andthe data exchange cost among vehicles, we formulate the problem as amixed-integer non-linear programming problem, which is Np-hard. Moreover, theconstraint associated with preserving job structures poses addressing thesubgraph isomorphism problem, that further complicates the algorithm design.Motivated by which, we propose an efficient decoupled approach by separatingthe template (feasible mappings between components and vehicles) searching fromthe transmission power allocation. For the former, we present an efficientalgorithm of searching for all the subgraph isomorphisms with low computationcomplexity. For the latter, we introduce a power allocation algorithm byapplying convex optimization techniques. Extensive simulations demonstrate thatthe proposed approach outperforms the benchmark methods considering variousproblem sizes.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×