A New Active Queue Management Algorithm Based on Neural Networks Pi
نویسندگان
چکیده
Active Queue Management (AQM) applies a suitable control policy upon detecting congestion in networks. In this paper, an adaptive Proportional-Integral (PI) controller based on Artificial Neural Networks (ANN) is applied to AQM for the objective of congestion avoidance and control in middle nodes. The proposed controller is simple and can be easily implemented in high-speed routers. Neural Network PI (NNPI) dynamically adapts its parameters with respect to changes in the system. It is anticipated that this results in better response compared to linear controllers due to the nonlinear nature of NNPI. We simulated our method in ns2 and compared its performance with the conventional PI controller. The simulation results show NNPI yields better performance. Copyright © 2005 IFAC
منابع مشابه
Direct Power Control With NSTSM Algorithm for DFIG Using SVPWM Technique
The paper presents a super-twisting sliding mode (STSM) regulator with neural networks (NN) of direct power command (DPC) for controlling the active/reactive power of a doubly-fed induction generator (DFIG) using a two-level space vector pulse width modulation (2L-SVPWM). Traditional DPC strategy with proportional-integral (PI) controllers (DPC-PI) has significantly more active/reactive power r...
متن کاملActive Queue Management in TCP Networks Based on Fuzzy-Pid Controller
We introduce a novel and robust active queue management (AQM) scheme based on a fuzzy controller, called hybrid fuzzy-PID controller. In the TCP network, AQM is important to regulate the queue length by passing or dropping the packets at the intermediate routers. RED, PI, and PID algorithms have been used for AQM. But these algorithms show weaknesses in the detection and control of congestion u...
متن کاملForecasting Stock Market Using Wavelet Transforms and Neural Networks: An integrated system based on Fuzzy Genetic algorithm (Case study of price index of Tehran Stock Exchange)
The jamor purpose of the present research is to predict the total stock market index of Tehran Stock Exchange, using a combined method of Wavelet transforms, Fuzzy genetics, and neural network in order to predict the active participations of finance market as well as macro decision makers.To do so, first the prediction was made by neural network, then a series of price index was decomposed by w...
متن کاملDesign of H� Congestion Controller for TCP Networks Based on LMI Formulation
In this paper, a state feedback H¥ controller is proposed in order to design an active queue management (AQM) system based on congestion control algorithm for networks supporting TCP protocols. In this approach, the available link bandwidth is modeled as a time-variant disturbance. The purpose of this paper is to design a controller which is capable of achieving the queue size and can guarantee...
متن کاملA New Pi and Pid Control Design Method and Its Application to Active Queue Management of Tcp Flows
A NEW PI AND PID CONTROL DESIGN METHOD AND ITS APPLICATION TO ACTIVE QUEUE MANAGEMENT OF TCP FLOWS Deniz Üstebay M.S. in Electrical and Electronics Engineering Supervisor: Prof. Dr. Hitay Özbay June 2007 PID controllers are continuing to be used in many control applications due to their simple structures. Design of such controllers for unstable systems with time delays is an active research are...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2005