Search

This paper studies the congestion control problem of multiple bottleneck networks. First, this paper regards the routing nodes of the network as multi-agents, and considers its own dynamic process. For the first time, the time-varying characteristics of the number of sessions in the network transmission process and the mutual influence between nodes are considered into the model, and a queue utilization rate switching model is established to describe the dynamics of a single agent. Then, considering the boundedness of the control input, combined with the prescribed performance technique, two congestion controllers are designed by using the backstepping method according to whether the controller depends on the number of sessions. The designed congestion controllers improve the robustness of the TCP/AQM network system in the case of frequent and time-varying sessions in multiple bottleneck networks. In addition, the designed controllers can stabilize different routing nodes of multiple bottleneck networks under different queue lengths, thereby improving the utilization efficiency of the network. Finally, the effectiveness and superiority of the designed controllers are verified by simulation. • TCP/IP network node queue utilization switching model is built. • The controller design takes into account the input saturation characteristics. • Two prescribed performance congestion control algorithms are designed. [ABSTRACT FROM AUTHOR]

This article investigates the adaptive fuzzy prescribed time H ∞ congestion control problem for a class of high speed transmission control protocol/active queue management (HSTCP/AQM) systems subject to asymmetric input saturation and external disturbance. Firstly, a saturation-based finite-time performance function (SFTPF) is constructed by introducing a nonnegative modified signal associated with input saturation errors into user-defined performance boundaries, which features the capability of ensuring the user-appointed tracking performance when no saturation exists and degrading it once saturation occurs, thereby effectively increasing the flexibility and reliability of the control scheme. Then, by designing a novel shifting function, the condition that the initial value of the queue tracking error remains within that of SFTPF is relaxed, which avoids the SFTPF being redeveloped for different initial queue tracking errors. Thereafter, an auxiliary system and H ∞ control theory are employed to eliminate the adverse effect caused by input saturation and external disturbance, respectively. Moreover, through the Lyapunov stability analysis, it is proven that, under our proposed adaptive fuzzy congestion control scheme, all the closed-loop signals are bounded, and the connection between input saturation and performance-related constraint is established. Finally, a set of simulation results are presented to demonstrate the feasibility and effectiveness of the designed congestion controller. • A new shifting function is developed to relax the initial condition restriction. • An error transformation function is designed to solve asymmetric constrained issues. • The performance constraint boundaries can be resiliently enlarged and recovered. • An adaptive fuzzy practical prescribed time congestion control scheme is proposed. [ABSTRACT FROM AUTHOR]

In this paper, a performance constraint control problem is considered for Transmission Control Protocol/Active Queue Management (TCP/AQM) network with external disturbance and input saturation. On the basis of backstepping-like design procedure and fuzzy approximation technique, an adaptive fuzzy controller with prescribed constraint is achieved to ensure that the transient and steady state performances of the tracking errors can be satisfied. The stability analysis proves that all the signals in the closed-loop system are semi-globally, uniformly and ultimately bounded. The simulation results clarify the feasibility and effectiveness of the proposed approach. • A new generalized AIMD network model with input saturation and external disturbances is proposed by combining the two existing models for Transmission Control Protocol/Active Queue Management (TCP/AQM) network. • A novel congestion control method is presented to solve a tracking problem by using funnel control, adaptive backstepping, and fuzzy logic systems. • The steady and transient state performances of the tracking error can satisfy the design requirements. [ABSTRACT FROM AUTHOR]

The congestion control problem of TCP network systems with user datagram protocol (UDP) flows is investigated in this paper. A nonlinear TCP network model with strict-feedback structure is first established. The unknown UDP flow is regarded as the external disturbance, and the maximum UDP flow is calculated by using the minimax approach. And then, a congestion control algorithm is proposed by using the adaptive backstepping approach. Meanwhile, the adaptive law is employed to estimate the unknown link capacity. The design of the adaptive law is to introduce a parameter mapping mechanism to limit the parameter identification range to a specified interval, thereby improving the estimation efficiency of the parameters. Furthermore, a state-feedback congestion controller is presented to make sure that the output of the system tracks the desired queue. The simulation results show the superiority and feasibility of the proposed method. • The UDP flow is addressed by the minimax method instead of H ∞ control. • A novel congestion algorithm is presented to solve an adaptive tracking problem. • The steady-state error of the closed-loop system satisfies the design requirements. [ABSTRACT FROM AUTHOR]

Inspired by the prescribed performance control (PPC), a new performance function, called finite-time performance function (FTPF), is first defined in this paper, by which a novel finite-time control design process is introduced. This work is also the first to solve the finite-time control issue for a class of transmission control protocol/active queue management (TCP/AQM) networks. Meanwhile, with the aid of FTPF, PPC and neural networks, a new adaptive practically pre-assigned finite-time controller is designed, which guarantees that the queue length q (t) tracks the desired queue q ref in finite-time and all the signals of the closed-loop system are semi-globally practical finite-time stable (SGPFS). Finally, simulation results are shown to illustrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

An H ∞ congestion tracking control problem for nonlinear transmission control protocol/active queue management (TCP/AQM) network is proposed in this paper. Inspired by the existing network models, a modified TCP/AQM network model with external disturbance and modelling uncertainty is first introduced. And then, a novel function for the desired queue length is defined for the first time, which can simulate the changes in the queue length during a day. Next, a novel AQM scheme is presented to control network congestion by combining H ∞ theory and integral backstepping technique. The proposed method guarantees that the better tracking performance of the queue is achieved, and all the signals are asymptotically stable in the closed-loop system in probability. Finally, the effectiveness of the presented algorithm is tested by simulation results. [ABSTRACT FROM AUTHOR]

This paper extends the well-known control method, prescribed performance control (PPC), to network congestion control problems. An adaptive H ∞ tracking problem for Transmission Control Protocol/Active Queue Management (TCP/AQM) system with external disturbance is studied. Firstly, a modified network model is given. And then, the model is changed to an equivalent error model by using error transformation. Next, to solve the network congestion problem, prescribed performance, backstepping technique, adaptive control and H ∞ control are combined to design a congestion controller. Due to the use of prescribed performance, the controller can guarantee both the transient and steady state performance of the system. Meanwhile, the output of the system can track the desired queue, and unknown link capacity can be estimated. Finally, a simulation result is shown to clarify the feasibility and effectiveness of proposed approach. [ABSTRACT FROM AUTHOR]

A novel network congestion algorithm is introduced for Transmission Control Protocol/Active Queue Management (TCP/AQM) system in this paper. The established TCP/AQM system is more accurate and general. Moreover, an adaptive congestion controller is designed by virtue of the Barrier Lyapunov Function (BLF), backstepping-like and Neural Networks (NNs) approximation techniques, by which the transient and steady-state performances on the tracking error can be pre-given and other signals of the closed-loop system also are verified to be semi-globally, uniformly and ultimately bounded. Finally, a comparison example is considered to demonstrate the feasibility and superiority of the presented scheme. [ABSTRACT FROM AUTHOR]