Abstract: Based on the existing control model for network congestion, a class of networked system with time-varying parameter uncertainties is considered with the aim of dealing with the problem of prevalent parameter uncertainties. Linearizing a previously developed nonlinear dynamic model of TCP, a linear time-varying system with parameter uncertainties is obtained to represent the intrinsic nonlinear characteristic more accurately. Then, applying the Lyapunov stability theory and linear matrix inequalities (LMIs) approach to this linear model, a convex stability condition is obtained and, simultaneously, a guaranteed cost controller design method is proposed to solve the problem of congestion control. The theoretical analysis and simulation results showed that the method proposed can guarantee some given performance indices in addition to the better robustness and faster convergence than some others.