Intelligent Multi-Link Selection for IEEE 802.11be Using Reinforcement Learning with Radiation Pattern Awareness

The IEEE 802.11be (Wi-Fi 7) standard introduces Multi-Link Operation (MLO) to enhance network efficiency by enabling simultaneous data transmission across multiple frequency bands. However, optimal link selection in dynamic wireless environments remains a significant challenge due to fluctuating signal conditions, interference, and the impact of antenna radiation patterns. Traditional heuristic-based approaches often fail to adapt effectively to these variations, leading to suboptimal performance. This paper proposes an intelligent Multi-Link selection framework based on Reinforcement Learning (RL), incorporating radiation pattern awareness to enhance decision-making. The proposed model leverages both omnidirectional and directional beamforming characteristics to dynamically select the most efficient link, considering real-time network conditions such as signal strength, interference, and spatial coverage. By integrating radiation pattern information into the RL training process, the system optimizes link selection while minimizing latency and improving throughput. Simulation results demonstrate that the proposed RL-based approach can select links based on various factors and dynamically choose the optimal link in real-time. This study highlights the potential of RL-driven Multi-Link selection strategies in advancing next-generation Wi-Fi networks and adaptive wireless communication systems.