Title: Analysis of Early Traffic Processing and Comparison of Machine Learning Algorithms for Real Time Internet Traffic Identification Using Statistical Approach
Abstract: In modern Internet, different protocols generate numerous traffic types with distinct service requirements. Therefore the Internet traffic identification plays an important role to improve the network performance as part of network measurement and network management task. Primarily well-known port based method was used. But latest services uses random and uncertain port numbers reduces the accuracy of identification. Consequently "payload based approach" also known as "deep packet inspection", used but still resulted less accuracy and required huge operational resources and are exposed to encrypted traffic flows. The recent techniques classify the application protocol based on statistical characteristics at packet level using network flow-based approach. Dealing with several datasets and millions of transaction of packets needs the use of Machine learning techniques for classification and identification of traffic. Our research shows the classification accuracy up to 99.7929%. In this paper we propose the statistical feature based approach for real-time network traffic classification. We compared the performance of three machine learning algorithms for the same. This mechanism of real time protocol identification confirms improved performance and reduced complexity.
Publication Year: 2014
Publication Date: 2014-01-01
Language: en
Type: book-chapter
Indexed In: ['crossref']
Access and Citation
Cited By Count: 9
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot