A deep learning‐based framework to identify and characterise heterogeneous secure network traffic

The evergrowing diversity of encrypted and anonymous network traffic makes management more formidable to manage the traffic. An intelligent system is essential analyse identify accurately. Network needs such techniques improve Quality Service ensure flow secure However, due usage non-standard ports encryption data payloads, classical port-based payload-based classification fail classify secured To solve above-mentioned problems, this paper proposed an effective deep learning-based framework employed with flow-time-based features predict heterogeneous best. state-of-the-art machine learning strategies (C4.5, random forest, K-nearest neighbour) are investigated for comparison. 1D-CNN model achieved higher accuracy in classifying In next step, characterises major categories (virtual private traffic, onion router plain traffic) into several application types. experimental results show effectiveness feasibility framework, which yields improved predictive power compared analysis.