Optimal Design of Resource Discovery-Allocation- Transmission-Offloading Strategy in Mobile Edge Computing
- Resource Type
- Conference
- Authors
- Guo, Rongzong; Liu, Yi; Lin, Ziqiong; Zheng, Yifeng; Zhang, Wenjie; Yang, Jingmin
- Source
- 2022 International Conference on Machine Learning, Cloud Computing and Intelligent Mining (MLCCIM) MLCCIM Machine Learning, Cloud Computing and Intelligent Mining (MLCCIM), 2022 International Conference on. :6-12 Aug, 2022
- Subject
- Computing and Processing
Energy consumption
Cloud computing
Multi-access edge computing
5G mobile communication
Decision making
Machine learning
Programming
mobile edge computing
resource discovery
resource allocation
computation offloading
dynamic programming
- Language
Mobile Edge Computing (MEC) is a key technology of IoTs and 5G networks and presents many challenges (such as resource discovery, resource allocation, computation offloading and transmission power design) that need to be addressed. Previous works studying on improving the performance of MEC system may only optimize these issues separately or just jointly optimized part of them. In this paper, we propose a Discovery-Allocation-Transmission-Offloading (DATO) strategy by taking all these aspects into consideration with discovery order to minimize energy consumption. Discovery strategy indicates when to suspend edge node (EN) discovery and perform computation offloading. Transmission strategy specifies the optimal power used to transmit the data. The offloading strategy tells the portion of data offloaded to the EN for execution. The discovery order demonstrates the sequence of EN searching. This problem is formulated as a stochastic sequential decision-making problem and Dynamic Programming (DP) is used to achieve the optimal scheme. Numerical results show the effectiveness of our proposed strategy.