A traffic recognition based on artificial neural networks in SDN systems

Autor:	Ho-Ting Wu, Yu-Shan Jhou, Kai-Wei Ke
Rok vydání:	2020
Předmět:	Identification (information) Statistical classification ComputingMethodologies_PATTERNRECOGNITION Correctness Artificial neural network Network packet Computer science business.industry Quality of service Deep learning Artificial intelligence business Convolutional neural network
Zdroj:	ICCE-TW
Popis:	This paper presents a traffic recognition, the basis of QoS, based on artificial neural networks in SDN systems. To the design methodology of systems, this paper applied successive semi-supervised learning and based upon Convolutional Neural Network (CNN) to enabling packet identification and determine the traffic type of a flow. The resulting reveals up to ninety-five percent of correctness of given/labelled flow recognition and a ninety-two percent of correctness for unknown flows.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::cac645ad6c9d676456d9cd5c5b6e7909 https://doi.org/10.1109/icce-taiwan49838.2020.9258056 Zobrazit plný text záznamu