Autonomous On-Device Protocols: Empowering Wireless with Self-Driven Capabilities
Hannaneh Barahouei Pasandi, Tamer Nadeem · 2024
This paper presents a study on applying on-device machine learning (ML) algorithms to enhance MAC layer protocols in wireless communications. It focuses on the MU-MIMO Grouping algorithm and explores the benefits of executing ML models directly on devices such as computers, smartphones, and IoT devices. This approach promises improved speed, privacy, security, and adaptability in dynamic networks. The paper evaluates the effectiveness of this strategy in Wi-Fi and Mas-sive MIMO scenarios, demonstrating significant system capacity enhancement, latency reduction, and improved user experience. Additionally, it examines the interaction between on-device ML and changing network environments, underscoring the method's adaptability and robustness. This research represents a significant advancement in MAC layer protocols using on-device ML and may inspire future innovations in wireless networks.