Adaptive finite-time sliding-mode tracking control for congestion of TCP networks under perturbations

This paper addresses the tracking control problem of congestion in TCP networks. Firstly, the system modeling of TCP is provided, followed by the presentation of state variables in the presence of both matched and mismatched perturbations. The system is affected by complex matched and mismatched uncertainties, which degrade tracking performance and increase the system's difficulty in control. To address this challenge, the paper explores nonsingular terminal sliding mode control, and the finite-time stability of the closed-loop system is demonstrated. Additionally, based on the proposed sliding variable, the queue states converge to fixed-time, and this estimation is dependent solely on the control parameters, not on initial queue conditions. To mitigate the chattering problem, an upper bound of disturbances is estimated through adaptive control laws. The Lyapunov analysis is employed to validate all theoretical results. Finally, numerical simulations are presented to illustrate the effectiveness of the proposed approach.