Computation Offloading and Resource Allocation in Vehicular Networks Based on Dual-Side Cost Minimization
Top Cited Papers
- 23 November 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 68 (2), 1079-1092
- https://doi.org/10.1109/tvt.2018.2883156
Abstract
The proliferation of smart vehicular terminals (VTs) and their resource hungry applications imposes serious challenges to the processing capabilities of VTs and the delivery of vehicular services. Mobile Edge Computing (MEC) offers a promising paradigm to solve this problem by offloading VT applications to proximal MEC servers, while TV white space (TVWS) bands can be used to supplement the bandwidth for computation offloading. In this paper, we consider a cognitive vehicular network (CVN) that uses the TVWS band, and formulate a dual-side optimization problem, to minimize the cost of VTs and that of the MEC server at the same time. Specifically, the dual-side cost minimization is achieved by jointly optimizing the offloading decision and local CPU frequency on the VT side, and the radio resource allocation and server provisioning on the server side, while guaranteeing network stability. Based on Lyapunov optimization, we design an algorithm called DDORV to tackle the joint optimization problem, where only current system states, such as channel states and traffic arrivals, are needed. The closed-form solution to the VT-side problem is obtained easily by derivation and comparing two values. For MEC server side optimization, we first obtain server provisioning independently, and then devise a continuous relaxation and Lagrangian dual decomposition based iterative algorithm for joint radio resource and power allocation. Simulation results demonstrate that DDORV converges fast, can balance the cost-delay tradeoff flexibly, and can obtain more performance gains in cost reduction and as compared with existing schemes.Keywords
Funding Information
- National Natural Science Foundation of China (61701399)
- Science and Technology Innovation Team of Shaanxi Province (2017KCT-30-02)
This publication has 33 references indexed in Scilit:
- Edge Computing: Vision and ChallengesIEEE Internet of Things Journal, 2016
- WhiteFi Infostation: Engineering Vehicular Media Streaming With Geolocation DatabaseIEEE Journal on Selected Areas in Communications, 2016
- Intelligent Energy and Traffic Coordination for Green Cellular Networks With Hybrid Energy SupplyIEEE Transactions on Vehicular Technology, 2016
- Vehicular Fog Computing: A Viewpoint of Vehicles as the InfrastructuresIEEE Transactions on Vehicular Technology, 2016
- Leakage-Aware Dynamic Resource Allocation in Hybrid Energy Powered Cellular NetworksIEEE Transactions on Communications, 2015
- DREAM: Dynamic Resource and Task Allocation for Energy Minimization in Mobile Cloud SystemsIEEE Journal on Selected Areas in Communications, 2015
- Enabling coexistence of cognitive vehicular networks and IEEE 802.22 networks via optimal resource allocationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Collaborative Task Execution in Mobile Cloud Computing Under a Stochastic Wireless ChannelIEEE Transactions on Wireless Communications, 2014
- When vehicles meet TV white space: A QoS guaranteed dynamic spectrum access approach for VANETPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2014
- Stochastic Network Optimization with Application to Communication and Queueing SystemsSynthesis Lectures on Communication Networks, 2010