TY  - JOUR
T1  - Multiple Access Control Protocol using Deep-Reinforcement Learning in Heterogeneous Wireless Networks
AU  - Kim, Do-won 
AU  - Shin, Kyung-seop 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2024
DA  - 2024/1/1
DO  - 10.7840/kics.2024.49.1.88
KW  - Heterogeneous Network
KW  - Multiple Access Control
KW  - Reinforcement Learning
KW  - Packet Collisions
AB  - With the increasing number and diversity of mobile communication devices, the efficient allocation of limited frequency bands has become a critical concern. In heterogeneous network environments, where multiple devices coexist within a single network, the application of different Multiple Access Control (MAC) protocols to each device leads to inevitable collisions using conventional methods. In this paper, we propose MAC protocol based on reinforcement learning, aiming to achieve efficient coexistence in such heterogeneous networks. By utilizing reinforcement learning, our proposed protocol mitigates collisions in data transmission, even in scenarios with hardness of information exchange between devices and experimental results demonstrate performance improvements in mixed-network environments.