Eman Alqudah awarded best paper: GCN-Driven Reinforcement Learning for Probabilistic Real-Time Guarantees in Industrial URLLC

Student Name: Eman Alqudah

Academic Department: Electrical and Computer Engineering

Advisor: Ashfaq Khokhar

Accomplishment: Best Paper Award – GCN-Driven Reinforcement Learning for Probabilistic Real-Time Guarantees in Industrial URLLC

Description: Ensuring packet-level communication quality is vital for ultra-reliable, low-latency communications (URLLC) in large-scale industrial wireless networks. In this paper, we enhance the Local Deadline Partition (LDP) algorithm by introducing a Graph Convolutional Network (GCN) integrated with a Deep Q-Network (DQN) reinforcement learning framework to achieve improved interference coordination in multi-cell, multi-channel networks.

Unlike LDP’s static priority mechanism, the proposed GCN-DQN model dynamically learns link priorities based on real-time traffic demand, network topology, remaining transmission opportunities, and interference patterns. The GCN captures spatial dependencies, while the DQN enables adaptive scheduling decisions through reward-guided exploration.

Simulation results show that our GCN-DQN model achieves mean SINR improvements of 179.6%, 197.4%, and 175.2% over LDP across three network configurations comprising 83, 163, and 324 wireless links, respectively. Additionally, the GCN-DQN model demonstrates mean SINR improvements of 31.5%, 53.0%, and 84.7% over our previous CNN-based approach across the same configurations.

This work provides a significant advancement in AI-enabled resource allocation for industrial URLLC systems. By integrating graph learning and reinforcement learning, the proposed approach enables probabilistic real-time guarantees with minimal computational overhead, offering a scalable and intelligent solution for complex 5G/6G industrial communication environments.

Reference website: Here