Energy aware scheduling using reinforcement learning for 802.15.4e Time-Slotted Channel Hopping
Abstract
Time Slotted Channel Hopping (TSCH) is a medium access control mode defined in IEEE 802.15.4e standard. This protocol is a crucial component of fourth-generation IoT applications, enabling efficient communication through synchronized time slots and channel hopping. By employing TSCH, IoT devices can achieve improved reliability, reduced interference, and increased network capacity. However, the energy consumption of IoT devices remains a significant challenge in large-scale deployments. This research introduces an energy-aware (EARL) schedule based on Reinforcement Learning (RL) for the 802.15.4e Time-Slotted Channel Hopping (TSCH) mode. The goal is to turn off slots in the 802.15.4e TSCH frame that are not highly utilized so as to conserve energy. By considering a predefined threshold, each node determines the slots that should be deactivated. This adaptive scheduling strategy allows the nodes to conserve energy effectively by minimizing unnecessary radio operations. Through extensive simulations and evaluations with simple and large-scale network configurations, the proposed energy-aware TSCH scheduling algorithm using Q-learning demonstrates promising results and is compared with the Orchestra protocol. This innovative approach reveals superior performance compared to Orchestra, achieving notably improved outcomes in both packet delivery rate and energy savings.