Dynamic Data Streams for Time-Critical IoT Systems in Energy-Aware IoT Devices Using Reinforcement Learning.

Area of Science:

• IoT Systems
• Energy Efficiency
• Machine Learning

Background:

• Energy-aware sensors are crucial for various monitoring applications.
• Extending sensor lifetime requires energy-efficient solutions for time-critical data.

Purpose of the Study:

• To propose a reinforcement learning (RL) based dynamic data stream solution for energy-aware IoT devices.
• To enhance data reliability and sensor battery lifetime in time-critical IoT systems.

Main Methods:

• Utilized the Q-Learning algorithm, a reinforcement learning technique.
• Developed a mechanism to dynamically adjust data transport rates based on available renewable energy.

Main Results:

• The proposed solution demonstrated an increase in transmitted data by up to 23%.
• Ensured continuous operation of IoT devices by balancing data transmission and energy availability.

Conclusions:

• Reinforcement learning effectively manages energy resources in IoT devices.
• The Q-Learning based approach enhances data transmission efficiency and device longevity.