Energy-Efficient AP Selection Using Intelligent Access Point System to Increase the Lifespan of IoT Devices

Area of Science:

• Computer Science
• Electrical Engineering
• Telecommunications

Background:

• Internet of Things (IoT) devices face energy consumption challenges in crowded environments.
• Collisions in wireless networks lead to packet retransmissions, increasing energy use and latency.
• Existing access point (AP) selection schemes do not adequately address energy efficiency and load balancing.

Purpose of the Study:

• To develop a novel energy-efficient AP selection scheme for IoT devices.
• To mitigate unbalanced load issues caused by biased AP connections.
• To enhance the overall performance of IoT devices in terms of energy consumption and latency.

Main Methods:

• Proposed an Energy and Latency Reinforcement Learning (EL-RL) model for AP selection.
• Analyzed collision probability in Wi-Fi networks to minimize retransmissions.
• Integrated average energy consumption and average latency into the AP selection process.

Main Results:

• Achieved a maximum improvement of 53% in energy efficiency.
• Reduced uplink latency by up to 50%.
• Extended the expected lifespan of IoT devices by 2.1 times compared to conventional methods.

Conclusions:

• The EL-RL model effectively enhances energy efficiency and reduces latency for IoT devices.
• The proposed scheme offers a significant improvement over traditional AP selection methods.
• This approach is crucial for optimizing IoT device performance in dense network environments.