فهرست نشریات علمی دسترسی آزاد ایران

In healthcare monitoring, where effective Medium Access Control (MAC) protocols are required to handle both periodic and emergency traffic, Wireless Body Area Networks (WBANs) are becoming increasingly significant. Conventional slot allocation techniques often lack adaptability to changing network conditions, which results in increased energy usage, longer wait times, and decreased dependability. To overcome these constraints, this study suggests a dynamic slot allocation mechanism for WBAN MAC protocols based on a Decision Tree. The Decision Tree Traffic Adaptive MAC (DTTA-MAC) approach allocates time slots according to real-time node conditions, ensuring that emergency traffic is prioritized while maintaining stable performance for normal traffic. A dataset was generated using Castalia and OmNeT++ to simulate realistic WBAN scenarios, incorporating parameters such as energy levels, latency, buffer size, throughput, and signal quality. Simulation results show that the proposed DTTA-MAC achieves up to approximately 34% reduction in energy consumption and about 89% reduction in average packet delay compared to existing MAC protocols, particularly under emergency traffic conditions. This work provides a portable and interpretable substitute for Reinforcement Learning (RL) techniques. To support real-world healthcare applications, future research will investigate expanding the model to larger networks, integrating it with cutting-edge machine learning techniques, and bolstering security and privacy.