We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
A New Scheme for Network Congestion Control Based on Modified Adaptive Smith Predictor.
- Authors
Sheng Xu; Minrui Fei; Xinhao Yang
- Abstract
In this paper, a simplified network model based on fluid flow theory and stochastic differential equations is derived. Based on the linearized TCP/AQM model, the modified adaptive strategy is applied to adjust the traditional Smith predictor for reducing the dependency on the mathematical model. Then, an Adaptive Modified Smith-PI controller (AMS-PI) is introduced as a new AQM scheme. According to historical data, this algorithm can dynamically modify the output, compensate for the influence of time delay, and reduce effectively the sensitivity to the actual model. The simulation results demonstrate that the performance of the proposed AMS-PI scheme are clearly superior to those of the PI scheme under both cases involving long delay and burst data flow.
- Subjects
PID controllers; STOCHASTIC differential equations; DATA flow computing
- Publication
International Journal of Simulation: Systems, Science & Technology, 2016, Vol 17, Issue 28, p1
- ISSN
1473-8031
- Publication type
Article
- DOI
10.5013/IJSSST.a.17.28.30