Packet Identification By Using Data Mining Techniques


Accurate internet traffic identification and classification are fundamental to numerous network activities, including network management and security monitoring, traffic modeling and network planning, accounting and Quality of Service provision. With the development of network, P2P as new generation of network technology is widely used. Starting from the first popular one (Napster), a number of new P2P based multimedia file sharing systems have been developed (FastTrack, eDonkey, Gnutella, Direct Connect, etc.). A fundamental types of networks architectures in today's world are Client/ server and Peer to Peer. A promising approach that has recently received some attention is traffic classification using machine learning techniques. The term data mining is used for methods and algorithms that allow analyzing data in order to find rules and patterns describing the characteristic properties of the data. The aim of this research is to classify traffic accuracy which can be achieved by using machine learning techniques such as K-Means and Birch algorithms. This system depends on the extracted attributes and then use it in the proposed system to distinguish all types of packets. The goal of system of packet identification is to detect the types of packets and identification of application usage and trends , also identification of emerging applications diagnosing anomalies is critical for both network operators and end user in term of data security and service availability.