Medium Access Control
Medium Access Control (MAC) protocols govern how multiple devices share a wireless communication channel, aiming to maximize efficiency and reliability. Current research heavily emphasizes data-driven approaches, employing machine learning techniques like reinforcement learning (including multi-armed bandits and deep reinforcement learning) and neural networks to optimize resource allocation and adapt to dynamic network conditions. This focus stems from the increasing complexity and heterogeneity of modern wireless networks, particularly in IoT and 5G/6G systems, where traditional MAC protocols struggle to maintain optimal performance. Improved MAC protocols will lead to more efficient and reliable wireless communication across a wide range of applications.
Papers
Multi-armed Bandit Learning for TDMA Transmission Slot Scheduling and Defragmentation for Improved Bandwidth Usage
Hrishikesh Dutta, Amit Kumar Bhuyan, Subir Biswas
Reinforcement Learning for Protocol Synthesis in Resource-Constrained Wireless Sensor and IoT Networks
Hrishikesh Dutta, Amit Kumar Bhuyan, Subir Biswas