Energy-Efficient Online Resource Management and Allocation Optimization in Multi-User Multi-Task Mobile-Edge...

Area of Science:

• Computer Science
• Electrical Engineering

Background:

• Mobile Edge Computing (MEC) offloads tasks from wireless devices (WDs) to MEC servers, crucial for the Internet of Things (IoT).
• Resource management in MEC is complex due to unpredictable task arrivals, channel conditions, and energy usage.
• Hybrid energy harvesting in WDs adds another layer of complexity to energy-efficient resource allocation.

Purpose of the Study:

• To develop an energy-efficient joint resource management and allocation (ECM-RMA) policy for multi-user, multi-task MEC systems.
• To minimize time-averaged energy consumption while ensuring data and energy queue stability.
• To achieve a controllable trade-off between energy consumption and delay.

Main Methods:

• Formulated a stochastic optimization problem to minimize energy consumption under queue stability constraints.
• Decomposed the problem into two deterministic sub-problems solvable via convex optimization and linear programming.
• Proposed the ECM-RMA algorithm, which operates without prior knowledge of stochastic processes.

Main Results:

• The ECM-RMA algorithm effectively reduces time-averaged energy consumption in MEC systems.
• The algorithm achieves an energy consumption-delay trade-off expressible as [O(1/V), O(V)], controlled by a weight V.
• Simulation results validated the theoretical analysis and the algorithm's effectiveness.

Conclusions:

• The proposed ECM-RMA policy offers an effective solution for energy efficiency in MEC systems.
• The algorithm provides a flexible mechanism to balance energy consumption and delay.
• This work contributes to optimizing resource management in energy-constrained IoT environments.