Application traffic classification using statistic signature

摘要: Networks today are becoming more complex and diverse because of the appearance new applications services. The importance, therefore, application-level traffic classification is increasing daily. Application-level has become a very popular area study. Although many proposals have been presented, including port-based, payload-based machine learning-based methods, method that can manage all not yet developed. More recently, methods based on statistical flow information studied. In this paper, we propose an methodology using statistic signature. Our creates signature payload size, transmission order, direction first N packets in flow, uses to classify application traffic. Then, verification system, prove feasibility our show its high accuracy.