Q-Learning-Driven Enhancement of Slotted ALOHA in IEEE 802.15.4 WSNs

Given the proliferation of connected devices and the prioritization of real-time data acquisition across various scenarios, enhancing the energy efficiency within Wireless Sensor Networks (WSNs) is of paramount importance. This work has focused on the IEEE 802.15.4 standard and addresses existing medium access control protocols such as CSMA or Slotted ALOHA and proposes refinements in the Slotted ALOHA protocol through incorporating techniques like Binary Exponential Backoff (BEB) and Q-learning. These enhancements have demonstrated to be promising in terms of average delay reduction, energy efficiency and bolstered network throughput. As it facilitates more efficient energy management it constitutes a robust alternative to conventional CSMA in WSN MAC sub-layer protocols.