[Paper Review] Resource Allocation in Full-Duplex Mobile-Edge Computing Systems with NOMA and Energy Harvesting

Zhaohui Yang, Jiancao Hou+1 more|arXiv (Cornell University)|Jul 31, 2018

Advanced Wireless Communication Technologies35 references4 citations

TL;DR

This paper proposes a resource allocation framework for full-duplex mobile-edge computing systems integrating NOMA and energy harvesting to minimize total system energy consumption. By reformulating the nonconvex problem into an equivalent convex one through recursive power modeling, the authors design an iterative algorithm that jointly optimizes power control, time scheduling, and computation offloading, achieving significant energy savings over conventional schemes.

ABSTRACT

This paper considers a full-duplex (FD) mobile-edge computing (MEC) system with non-orthogonal multiple access (NOMA) and energy harvesting (EH), where one group of users simultaneously offload task data to the base station (BS) via NOMA and the BS simultaneously receive data and broadcast energy to other group of users with FD. We aim at minimizing the total energy consumption of the system via power control, time scheduling and computation capacity allocation. To solve this nonconvex problem, we first transform it into an equivalent problem with less variables. The equivalent problem is shown to be convex in each vector with the other two vectors fixed, which allows us to design an iterative algorithm with low complexity. Simulation results show that the proposed algorithm achieves better performance than the conventional methods.

Motivation & Objective

• To address the energy consumption and spectral efficiency limitations in mobile-edge computing (MEC) systems due to user battery constraints.
• To integrate full-duplex (FD) operation, non-orthogonal multiple access (NOMA), and energy harvesting (EH) in a single MEC framework to enhance spectral and energy efficiency.
• To minimize total system energy consumption through joint optimization of power control, time scheduling, and computation offloading.
• To develop a low-complexity iterative algorithm for solving the resulting nonconvex optimization problem.

Proposed method

• Reformulates the original nonconvex problem into an equivalent problem with fewer variables using a recursive method to express uplink user power as a function of time, BS power, and offloaded data.
• Proves that the equivalent problem is convex in each of the three variables (power, time, data) when the other two are fixed, enabling iterative optimization.
• Designs an iterative algorithm that alternately optimizes the power vector, time vector, and offloading data vector, ensuring convergence with low complexity.
• Utilizes successive interference cancellation (SIC) at the base station to decode NOMA-multiplexed uplink signals from multiple users in each group.
• Models the full-duplex base station as simultaneously receiving uplink data and broadcasting downlink energy to support user operation.
• Applies perspective function and convexity analysis to prove the convexity of the reformulated problem in time and data variables, ensuring algorithmic stability.

Experimental results

Research questions

• RQ1 How can full-duplex operation be effectively combined with NOMA and energy harvesting in a mobile-edge computing system to improve spectral and energy efficiency?
• RQ2 What is the optimal joint allocation of power, time, and computation offloading that minimizes total system energy consumption in such a system?
• RQ3 Can the nonconvex resource allocation problem in FD-MEC with NOMA and EH be transformed into a convex problem for efficient solution?
• RQ4 What performance gain does the proposed algorithm achieve compared to conventional orthogonal multiple access (OMA) and non-FD schemes?

Key findings

• The proposed iterative algorithm achieves lower total energy consumption than conventional schemes, demonstrating superior energy efficiency in FD-MEC systems with NOMA and EH.
• The equivalent problem formulation reduces variable dimensionality, enabling efficient optimization through alternating convex optimization.
• The system performance improves when pairing users with strong and moderately strong channel gains, as opposed to pairing only the strongest users.
• The convexity of the reformulated problem in each variable set ensures convergence and low computational complexity of the iterative algorithm.
• Simulation results confirm that the proposed method significantly reduces energy consumption, especially in high-mobility or high-data-rate scenarios.