Energy-Aware Computation Offloading and Transmit Power Allocation in Ultradense IoT Networks
- 11 October 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 6 (3), 4317-4329
- https://doi.org/10.1109/jiot.2018.2875535
Abstract
To meet the surging demands on network throughput and spectrum resources arising with billions of IoT mobile devices (IMDs), ultra-dense networks are envisioned to be a promising technology, which gives rise to the so-called ultra-dense IoT networks. Meanwhile, with the constant emergence of new IoT applications, the conflict between computing-intensive applications and resource-constrained IMDs is increasingly prominent. By offloading computing-intensive tasks to the edge servers in close proximity, mobile-edge computing is expected as an effective solution to address this issue. However, computation offloading research in ultra-dense IoT networks is still scarce until now. Toward this end, we provide this paper to study the energy-aware task offloading problem with multiple edge servers in ultra-dense IoT networks, where diverse kinds of computation tasks are randomly requested by the IMDs and the computing resources at the edge servers change dynamically. An iterative searching based task offloading scheme is proposed as our solution, which jointly optimizes task offloading, computational frequency scaling, and transmit power allocation. Extensive numerical results demonstrate the superior performance of conducting task offloading among multiple edge servers, and corroborate the advantages of our scheme over existing works which either fixed computational frequency and transmit power, or neglected the impact of the IMDs’ residual battery.Keywords
Funding Information
- National Natural Science Foundation of China (61801360, 61771374, 61771373, 61601357)
- Fundamental Research Fund for the Central Universities (JB181507, JB171501, JB181506, JB181508)
- China 111 Project (B16037)
This publication has 55 references indexed in Scilit:
- Context Awareness over Transient CloudsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Context‐awareness over transient cloud in D2D networks: energy performance analysis and evaluationTransactions on Emerging Telecommunications Technologies, 2015
- New Perspectives on Future Smart FiWi Networks: Scalability, Reliability, and Energy EfficiencyIEEE Communications Surveys & Tutorials, 2015
- Optimal Cloudlet Placement and User to Cloudlet Allocation in Wireless Metropolitan Area NetworksIEEE Transactions on Cloud Computing, 2015
- Joint Optimization of Radio and Computational Resources for Multicell Mobile-Edge ComputingIEEE Transactions on Signal and Information Processing over Networks, 2015
- Stochastic Game-Theoretic Spectrum Access in Distributed and Dynamic EnvironmentIEEE Transactions on Vehicular Technology, 2014
- Research Directions for the Internet of ThingsIEEE Internet of Things Journal, 2014
- Solving the Maximum Cardinality Bin Packing Problem with a Weight Annealing-Based AlgorithmPublished by Springer Science and Business Media LLC ,2009
- Fine-grained energy profiling for power-aware application designACM SIGMETRICS Performance Evaluation Review, 2008
- Processor design for portable systemsJournal of Signal Processing Systems, 1996