Key Technologies Research Of Network Traffic Measurement And Analysis

With the rapid development of network technology and wider applications of Internet services, the Internet is getting increasingly bulky. The supplies of the Internet and requirements of customers hence are becoming more and more diverse and complex, so there are more requirements of network performance and Quality of Service being brought forward. The network traffic can fully reflect complex dynamic characteristics of Internet and is also a dominant factor of network characteristics, so it is an important way to improve network performance by traffic control,shaping and dispatch. It is also important for network modeling and performance analysis.An essential task of network performance research based on traffic is traffic measurement, modeling and analysis which are bases of traffic control and a necessary way of performance analysis and quality of service. It is necessary to investigate the relationship between traffic and network performance parameters because network traffic plays an important role in network performance research and its characteristics exhibit by network performance parameters such as packet loss, delay and delay jitter。This is the work of traffic measurement, modeling and analysis. There are two problems: how traffic influences performance parameters and how performance parameters reflect traffic characteristics. The paper investigates the relationship between self-similar traffic and packet loss, and develops a method for cross-traffic inferring using delay jitter.This paper first introduces the aspects of network performance research based on traffic measurement, modeling and analysis and its state-of-the-art , secondly summarizes then the concept, models and analysis tools of self-similar traffic, and analyzes Scaling behavior of packet loss with self-similar traffic input by wavelets method, thirdly introduces Hidden Markov model and its applications on network performance research, and then explores the cross-traffic inferring technology and the disadvantages of existing methods. After that the paper develops a new method for cross-traffic inferring based on delay jitter measurement, proves its correctness by experiments, and applies it to self-similar traffic background and real traffic trace to investigate its availability,...