Thien Hieu Hoang, Chuyen T. Nguyen, Tri Nhu Do, and Georges Kaddoum
IEEE Transactions on Communications, Jan. 2024
In this paper, we provide:
- Introduction of a new uplink-downlink stochastic mobile edge computing (MEC) system, considering the dynamic characteristics of small cell base stations (SBSs) and user equipments (UEs).
- Development of a radio resource management strategy focusing on user association, channel allocation, and power allocation, incorporating time, energy, and achievable rate into the utility function.
- Identification of the problem as a Mixed Integer Nonlinear Program (MINLP), acknowledged as NP-hard.
- Proposal of a nature-inspired optimization framework to simplify the complexity, adaptable for subproblems in various settings and integrable with Whale Optimization Algorithm (WOA), Improved Whale Optimization Algorithm (IWOA), and Particle Swarm Optimization (PSO).
- Mathematical and numerical analysis showing the proposed algorithms' ability to reach near-optimal solutions with minimal optimality gaps.
- Numerical results highlighting the strengths and weaknesses of the proposed algorithms for resource management in MEC systems, and evaluating the impact of stochastic characteristics on MEC performance.