State estimation for a TCP/IP network using terminal sliding-mode methodology
Long Xu, Xinghuo Yu, Yong Feng, Fengling Han, Jiankun Hu, Zahir Tari · IECON 2017 - 43rd Annual Conference of the IEEE Industrial Electronics Society · 2017
Recently, the state estimation issue of a TCP/IP network has attracted much attention from different communities. In this paper, a terminal sliding-mode observer (TSMO) is proposed based on a fluid-flow model of a TCP/IP network to estimate traffic flow states. A novel control strategy is proposed to fasten the convergence of the estimation error for average congestion window (ACwnd). Furthermore, a continuous control strategy is directly used to estimates the flooding rate of additional traffic flow (ATF). The efficacy of the proposed TSMO is verified by a numerical simulation implementations via the networking simulator NS-2.