ACoCo: An Adaptive Congestion Control Approach for Enhancing CoAP Performance in IoT Network

The Industrial Internet of Things (IIoT) requires the real-time transmission of critical data to ensure functionality and prevent hazardous situations. However, current data transmission scheduling methods in 6TiSCH networks do not efficiently handle heterogeneous traffic based on its criticality and performance requirements, potentially leading to violations of timing limits. To address this issue, this paper proposes ACoCo, an Adaptive Congestion Control approach for CoAP that uses reinforcement learning techniques to dynamically adapt congestion control parameters based on real-time network conditions, node behaviors, and traffic patterns. Simulation results demonstrate ACoCo's effectiveness in reducing end-to-end transaction delay and improving transaction delivery ratio under congested network conditions, providing valuable insights for IoT network optimization and design. ACoCo operates effectively within the 6TiSCH network architecture, taking into account the scheduling function and communication requirements of the network.