Research Area:  Fog Computing
The diverse applications and high-quality services in smart cities have led to the geographical unbalance of computation requirements. Traditional centralized cloud computing services and the massive migration of computing tasks result in the increase of network delay and the aggravation of network congestion. Deploying fog nodes at the network edge has become an effective way to improve the quality of service. However, the dynamic requirements and application in various scenarios still challenge the network, resulting in the geographical unbalance of computing resource demands. Nowadays, the computing resources of on-board computers and devices in the Internet of Vehicles are abundant enough to mitigate the geographical unbalance in computing power demand. The efficient usage of the natural mobility of constantly moving vehicles to solve the above-mentioned problems remains an urgent need. In this paper, a vehicle mobility-based geographical migration model of the vehicular computing resource is established for the fog computing-enabled smart cities. The vehicle as a service framework takes the full advantage of the unbalance and randomness of vehicular computing resource and improves the flexibility of traditional cloud computing architecture. An incentive scheme that affects the vehicle path selection through resource pricing is proposed to balance the resource requirements and to geographically allocate computing resources. The simulation results indicate that the advantages and efficiency of the proposed scheme are significant.
Author(s) Name:  Siyi Liao; Jianhua Li; Jun Wu; Wu Yang; Zhitao Guan
Journal name:  IEEE Access
Publisher name:  IEEE
Volume Information:  ( Volume: 7) Page(s): 8726 - 8736
Paper Link:   https://ieeexplore.ieee.org/abstract/document/8598848