Deep progressive reinforcement learning-based flexible resource scheduling framework for IRS and UAV-assisted MEC system

TL;DR

This paper introduces a deep progressive reinforcement learning framework for optimizing resource scheduling in IRS and UAV-assisted MEC systems, effectively reducing energy consumption in dynamic scenarios.

Contribution

It proposes a multi-task deep reinforcement learning agent with progressive adaptation and a taboo search component for efficient, real-time resource scheduling in complex MEC systems.

Findings

Outperforms existing methods in energy efficiency.

Achieves real-time scheduling in dynamic UAV scenarios.

Demonstrates robustness to varying UAV numbers.

Abstract

The intelligent reflection surface (IRS) and unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) system is widely used in temporary and emergency scenarios. Our goal is to minimize the energy consumption of the MEC system by jointly optimizing UAV locations, IRS phase shift, task offloading, and resource allocation with a variable number of UAVs. To this end, we propose a Flexible REsource Scheduling (FRES) framework by employing a novel deep progressive reinforcement learning which includes the following innovations: Firstly, a novel multi-task agent is presented to deal with the mixed integer nonlinear programming (MINLP) problem. The multi-task agent has two output heads designed for different tasks, in which a classified head is employed to make offloading decisions with integer variables while a fitting head is applied to solve resource allocation with continuous variables. Secondly, a progressive scheduler is introduced to adapt the agent to the varying number of UAVs by progressively adjusting a part of neurons in the agent. This structure can naturally accumulate experiences and be immune to catastrophic forgetting. Finally, a light taboo search (LTS) is introduced to enhance the global search of the FRES. The numerical results demonstrate the superiority of the FRES framework which can make real-time and optimal resource scheduling even in dynamic MEC systems.