Adaptive congestion control in IoT networks: Leveraging one-way delay for enhanced performance

Area of Science:

• Computer Science
• Network Engineering
• Internet of Things (IoT)

Background:

• Exploding IoT device numbers generate vast data, risking performance degradation due to network congestion.
• Efficient congestion management is crucial for maintaining IoT network performance and reliability.

Purpose of the Study:

• To introduce and evaluate a novel One-Way-Delay (OWD)-based congestion control (CC) method for IoT networks.
• To enhance IoT device performance by mitigating network congestion through delay estimation and traffic adjustment.

Main Methods:

• Developed a congestion control method based on monitoring One-Way-Delay (OWD) across the communication path.
• Adjusted network traffic dynamically based on estimated data transmission delays.
• Conducted comparative analysis against established TCP variants (BBR, Cubic, HTCP, New Reno).

Main Results:

• The proposed OWD-based CC method demonstrated superior performance compared to existing TCP variants.
• Achieved average throughput improvements ranging from 4.1% to 22.7%.
• Maintained fairness in mixed-traffic environments and managed congestion effectively in complex network topologies.

Conclusions:

• The OWD-based CC method offers a more effective solution for IoT network congestion management.
• The approach improves network resource utilization, throughput, fairness, and reliability.
• This method is suitable for complex IoT network topologies and mixed-traffic conditions.