Network Traffic Flow Separation and Control Through a Hybrid ICA-Fuzzy Adaptive Algorithm

Authors:
Flávio Henrique Vieira;Lígia Maria Sousa;George E. Bozinis;Wesley F. Miranda;Charles Casimiro Cavalcante
Affiliations:
Escola de Engenharia Elétrica e de Computação (EEEC), Universidade Federal de Goiás (UFG),;Departamento de Engenharia de Teleinformática (DETI), Universidade Federal do Ceará (UFC),;Escola de Engenharia Elétrica e de Computação (EEEC), Universidade Federal de Goiás (UFG),;Escola de Engenharia Elétrica e de Computação (EEEC), Universidade Federal de Goiás (UFG),;Departamento de Engenharia de Teleinformática (DETI), Universidade Federal do Ceará (UFC),
Venue:
ICA '09 Proceedings of the 8th International Conference on Independent Component Analysis and Signal Separation
Year:
2009

Citing 6
Cited 0

Adaptive filter theory (2nd ed.)

Adaptive filter theory (2nd ed.)
An information-maximization approach to blind separation and blind deconvolution

Neural Computation
Independent Component Analysis: Principles and Practice

Independent Component Analysis: Principles and Practice
Enhancing network traffic prediction and anomaly detection via statistical network traffic separation and combination strategies

Computer Communications
Fuzzy adaptive predictive flow control of ATM network traffic

IEEE Transactions on Fuzzy Systems
Traffic models in broadband networks

IEEE Communications Magazine

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present a hybrid ICA-fuzzy adaptive algorithm for traffic flow separation and control in contemporary computer networks. Our approach is composed by an ICA Block corresponding to the gradient algorithm proposed by Bell and Sejnowski for the information maximization at the output of a neural network as well as a Fuzzy Control System Block. The ICA algorithm is used to separate the controllable to the non-controllable network traffic sources. Additionally, we developed a predictive fuzzy controller following the Takagi and Sugeno fuzzy modeling. The combination of blind separation and control algorithm is applied to real network traffic traces. Finally, we verify that the proposed ICA-fuzzy adaptive control algorithm yields prominent control performances for single buffer server network environments.