Mobile Computation Offloading in Mobile Edge Computing Based on Artificial Intelligence Approach: A Review and Future Directions

Mobile computation offloading (MCO) is one of the significant processes in mobile edge computing (MEC). MCO is a promising approach to contract with the restrictions in client devices by offloading resource-intensive tasks or at least a part of it to the nearby resource-rich servers in MEC. Since most of the MCO optimization models endeavor to solve an NP-hard problem, the approximation solutions with higher performance and lower complexity are proposed and evaluated in several studies. These solutions could be more optimized with the most recent developments in the Artificial Intelligence (AI) field such as machine learning (ML), and meta-learning (MTL). Lately, many AI techniques have been proposed to learn offloading policies through interacting with the MEC environment. This paper proposes a literature review for the recent mechanisms of ML-based and MTL-based MCO in MEC. A detailed study is proposed for the main issues, challenges, and future research direction in the field of MCO in MEC servers.