Active Queue Management Algorithm Based On Robust Control Is Studied | | Posted on:2014-01-31 | Degree:Master | Type:Thesis | | Country:China | Candidate:R Li | Full Text:PDF | | GTID:2248330395482520 | Subject:Control theory and control engineering | | Abstract/Summary: | PDF Full Text Request | | Computer network is the product of the combination of the computer technology and communication technology. The data transmitted over the network increased sharply as the Internet scale and application unprecedented growth. The congestion problem has aroused people’s close attention. The purpose of the Congestion control is to make the load not more than the network transmission capacity, and guarantee the network is stable and reliable.Active queue management (AQM) algorithm is the most effective method to solve the congestion problem. This paper analyses and summarizes the AQM algorithm systematically. At the sanme time, this paper analyzes the AQM algorithm based on control theory emphatically. This paper proposes several robust active queue management algorithms combined with robust control theory. The main work is as follows:(1) The robust control mechanism is introduced in congestion control. A discrete-time robust AQM controller based on H-infinity control theory is presented in this paper. The controller can make the closed-loop system asymptotically stable and meet the H-infinity performance index under the condition of the time-varying network parameters and link capacity disturbance.(2) We folly consider the unmodeled uncertainties influence based on considering the parameter uncertainties and link capacity disturbance. The unmodeled uncertainties are assumed as a one-dimensional wiener process. A stochastic robust AQM controller based on H-infinity control theory is presented in this paper. The controller can make the system asymptotically stable and meet the given performance index under the condition of the time-varying network parameters and link capacity disturbance.(3) Considering the parameter uncertainties, saturated input, link capacity disturbance and unmodeled uncertainties, then we design a Stochastic H-infinity robust controller based on observer combined with the robust control theory. The simulation shows that The controller can make the system asymptotically stable and meet the given performance index. | | Keywords/Search Tags: | Congestion Control, AQM algorithm, robust control, state observe, LMI | PDF Full Text Request | Related items |
| |
|