Conference Proceedings

An adaptive neuron AQM for a stable internet

Jinsheng Sun, Moshe Zukerman, IF Akyildiz (ed.), R Sivakumar (ed.), E Ekici (ed.), JC DeOliveira (ed.), J McNair (ed.)

NETWORKING 2007: AD HOC AND SENSOR NETWORKS, WIRELESS NETWORKS, NEXT GENERATION INTERNET, PROCEEDINGS | SPRINGER-VERLAG BERLIN | Published : 2007

Abstract

Recognizing that Internet congestion control is a complex nonlinear system, we propose here to use an intelligent controller to improve its stability and performance. In particular, we propose here a new, powerful, easy-to-configure and robust active queue management (AQM) scheme called adaptive neuron AQM (AN-AQM). We present extensive simulation results for AN-AQM, over a wide range of network conditions and scenarios, that demonstrate its attributes. We demonstrate its robustness in various realistic environments involving bursty HTTP connections and non-responsive UDP connections. Comparison with other AQM schemes has demonstrated the superiority of AN-AQM over well-known AQM schemes in ..

View full abstract

Grants

Awarded by Australian Research Council


Awarded by Natural Science Foundation of Jiangsu Province, China


Funding Acknowledgements

This work was jointly supported by grants from the Australian Research Council (Grant DP0559131), and the Natural Science Foundation of Jiangsu Province, China (No. BK2004132).