Research on Event-Triggered Active Queue Management Algorithms in TCP/IP Networks

As the scale of the network grows and the demand from users becomes larger, TCP/IP network congestion control has become increasingly important. In this paper, a series of event-triggered active queue management (AQM) algorithms are presented to improve bandwidth utilization and system stability. For linear TCP systems subject to disturbances and uncertainties, robust event-triggered controllers are co-designed using linear matrix inequality (LMI) techniques. To address unmeasurable system states, an observer-based AQM algorithm is presented, which demonstrates superior robustness and reduced computation frequency. Furthermore, a T-S fuzzy model-based event-triggered AQM scheme is presented to address network nonlinearity and time delay effects. Simulation results confirm the effectiveness of the proposed schemes in guaranteeing network stability with reduced resource utilization. This work promotes intelligent, resource-efficient congestion control mechanisms in complex network scenarios.