Smart IRS Resource Allocation for Multi-Access Edge Computing in IoT Networks

This paper presents an AI-driven framework that integrates Intelligent Reflecting Surfaces (IRS) with Multi-access Edge Computing (MEC) to optimize resource allocation in Next-Generation (Next-Gen) wireless networks. Designed to address communication challenges in environments with obstructed direct links, the system employs Uncrewed Aerial Vehicles (UAVs) equipped with IRS to facilitate efficient data transmission between mobile users and a Terrestrial Base Station (TBS) connected to a MEC server. The objective is to minimize network latency and energy consumption by jointly optimizing IRS configurations, computational resource distribution, and communication scheduling. To effectively address this complexity, a hierarchical heuristic approach is introduced, decomposing the problem into three subproblems: UAV trajectory optimization using a Grey Wolf Optimizer (GWO), task execution and scheduling via an Augmented Lagrangian Method (ALM), and IRS phase and UAV position refinement through a Multi-objective Differential Evolution Algorithm (MDEA). Simulation results demonstrate that the proposed approach significantly enhances data transmission efficiency, reduces latency, and optimizes energy consumption, highlighting the significant potential of IRS and MEC integration for addressing the stringent performance requirements of next-generation wireless networks, especially in complex and dynamic environments. For example, this framework is particularly suitable for urban smart city deployments where buildings frequently obstruct Line-of-Sight (LoS) links, as well as disaster recovery scenarios where damaged infrastructure hinders direct communication, requiring flexible and rapid deployment of UAV-assisted IRS to restore connectivity.