Reinforcement Learning (RL)-Based Energy Efficient Resource Allocation for Energy Harvesting-Powered Wireless Body...

Area of Science:

• Wireless Body Area Networks (WBANs)
• Energy Harvesting (EH) Technologies
• Resource Allocation Optimization

Background:

• WBANs enable continuous physiological monitoring but face battery limitations.
• Energy harvesting (EH) offers a sustainable power solution for WBANs.
• Efficient resource allocation is crucial for EH-WBANs to balance energy and Quality of Service (QoS).

Purpose of the Study:

• To maximize energy efficiency in EH-powered WBANs (EH-WBANs).
• To jointly optimize transmission mode, relay selection, time slots, and power.
• To address the energy constraints of individual sensors within the network.

Main Methods:

• Formulation of the energy efficiency problem as a discrete-time, finite-state Markov decision process (DFMDP).
• Development of a modified Q-learning (QL) algorithm to find the optimal allocation strategy.
• Decision-making by a central hub with limited network information.

Main Results:

• The proposed scheme effectively maximizes energy efficiency in EH-WBANs.
• The modified QL algorithm demonstrates a low computational complexity.
• Numerical results validate the proposed approach's effectiveness.

Conclusions:

• The developed resource allocation strategy enhances the longevity of EH-WBANs.
• The modified QL algorithm provides an efficient solution for complex optimization problems in WBANs.
• This research contributes to sustainable and reliable physiological monitoring systems.