Nonconvex integer optimal robust impulsive control strategy for first-order piecewise finite precision nonlinear random early detection algorithm

• Published in: ISA transactions Vol. 51; no. 3; pp. 439 - 445
• Main Authors: Yuk-Fan Ho, Charlotte, Wing-Kuen Ling, Bingo, Iu, Herbert H.C., Fernando, Tyrone L.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.05.2012; Elsevier
• Subjects: Adaptative systems; Algorithms; Applied sciences; Computer science; control theory; systems; Computer simulation; Computer systems and distributed systems. User interface; Control system synthesis; Control theory. Systems; Exact sciences and technology; First-order piecewise finite precision nonlinear dynamical model; Mathematical analysis; Mathematical models; Nonconvex integer optimal robust impulsive control; Nonlinearity; Optimal control; Optimization; Queues; Random early detection algorithm; Software; Strategy; First-order piecewise finite precision nonlinear dynamical model; Nonconvex integer optimal robust impulsive control; Random early detection algorithm; Computer simulation; Control synthesis; Impulse control; Equilibrium point; Robust control; Queue length; Optimal strategy; Power control; Optimal control; Traffic management; Non linear model; Force control; Optimal control (mathematics); Random early detection protocol; Non convex analysis
• ISSN: 0019-0578; 1879-2022; 1879-2022
• DOI: 10.1016/j.isatra.2011.11.007

There are two main contributions of this paper. First, this paper proposes a first-order piecewise finite precision nonlinear dynamical model for characterizing the average queue size of the random early detection (RED) algorithm. Second, this paper proposes a nonconvex integer optimal robust impulsive control strategy for stabilizing the average queue size. The objective of the control strategy is to determine the average queue size so that the average power of the impulsive control force is minimized subject to a constraint on the absolute difference between the actual average queue size and the theoretical average queue size at the equilibrium point. Computer numerical simulation results show that the proposed control strategy is effective and efficient for stabilizing the average queue size. ► We investigate random early detection (RED) algorithm for Internet traffic control. ► We develop a nonconvex integer optimal robust impulsive control strategy for chaotic systems. ► We develop a first-order piecewise finite precision nonlinear dynamical model for the RED algorithm.