A Reinforcement Learning Based Transmission Parameter Selection and Energy Management for Long Range Internet of...

Area of Science:

• Wireless Communication
• Internet of Things (IoT)
• Energy Efficiency in IoT

Background:

• LoRa-enabled IoT devices use Adaptive Data Rate (ADR) for transmission parameters.
• Accurate energy assessment of ADR parameter combinations is crucial for long-range applications.
• Balancing computational and transmission energy is key for effective physical parameter tuning.

Purpose of the Study:

• To provide a mathematical model for estimating energy consumption in LoRa end devices (EDs).
• To model LoRa energy optimization and transmission parameter selection as a Markov Decision Process (MDP).
• To enhance energy efficiency and reliability in LoRaWAN networks.

Main Methods:

• Developed a mathematical model for LoRa end device energy consumption estimation.
• Formulated the LoRa energy optimization and parameter selection as a Markov Decision Process (MDP).
• Evaluated the proposed method under various scenarios and compared it to LoRaWAN's default ADR.

Main Results:

• The proposed method significantly outperforms the standard LoRaWAN ADR mechanism in energy efficiency.
• The method allows LoRa end devices (EDs) to conserve more energy.
• Enhanced energy efficiency enables EDs to perform more transmissions and operate for longer durations.

Conclusions:

• The developed MDP-based approach offers superior energy optimization for LoRa-enabled IoT devices compared to standard ADR.
• This optimization leads to extended device operational life and increased data transmission capabilities.
• The findings are critical for improving the performance and sustainability of long-range IoT deployments.