PERFORMANCE ANALYSIS AND MODELLING OF IP TELEPHONY TRAFFIC

Abstract

The world is increasingly becoming IP Centric, one of the fastest growing internet application is the Internet Protocol Telephony (IPT), which allow voice transmission over Computer Network. This project focuses on IP Telephony (IPT) traffic as one of the recent internet applications. It investigates the characteristics of measured IPT traffic on both call and packet level, with explanations to the major factors that affect voice quality in an IP network. The result shows that internet voice traffic is being affected by the Codec used, Delay by the link, jitter, packet loss, which all contribute to the degradation of voice quality. To boost the quality of service (QoS) in an IP Telephony network, the frame work of this project was built on the modeling of an Adaptive filter using LMS (least mean square) algorithm, to filter noise in an IP Telephony network. The final error e is 0.4081 which shows that noise is totally cancelled from the speech signal. Assuming the final error signal is zero, in this situation the noise signal would be completely cancelled and the far user would not hear any form of unwanted sound in the original speech returned to them.