The Research And Test Of Voice Quality In VoIP System

VoIP (Voice over IP) is a new communication technology arisen in recent years. It is the result of the amalgamation by two mainstream technologies, one is packet network marked by IP, and the other is the integration of network service which is aimed at multimedia. Because of its low cost of communications, rich support of the business, it develops like a raging fire and is licking up the traditional PSTN phone market. However, real-time voice communications like VoIP needs some network transmission quality assurance, but the current internet can not meet this requirement, resulting its unsatisfactory of the current VoIP services on the stability and QOS. This has become the main obstacle and challenges to its further development as well as the major problem that IP phone technology has to deal with.How to carry out scientific and reliable voice quality measurement and evaluation is the key issue of improving VoIP quality. There are many factors which impact VoIP voice quality, mainly caused by the inherent characteristics of IP networks, including the network delay and jitter, packet loss, as well as out-of-order. These impairments will directly influence the communication. The parameters for impairment rely on jutter buffer to great extent. Therefore, we have to deal with jitter in VoIP communication system to improve its quality. The buffer should be neither too long nor too short to achieve the trade-off between the loss and delay. Most of the current communication system using adaptive jitter buffer algorithm to solve this problem.The main tasks and contributions of this paper are as follows:1. Studied the key fators that affect voice quality in VoIP communication and analysed the corresponding solutions.2. Illustrated the subjective and objective methods to evaluate voice quality of VoIP, and proposed a more efficency and accurate test method with the combination of PESQ and MOS.3. Explained how the jitter comes into being and what its influence to voice quality. We focus on study the exsiting 7 buffer algorithms, and based on this we designed an improved adaptive jitter buffer algorithm to better adapt to different network environment and optimize voice quality.4. We designed a simple but effective test system, and tested the above 8 algorithms. From the result, we can see that the new jitter buffer technology has better adaptiveness to different network environments, and it can reduce the packet loss, jitter and out-of-order to greater extent. This algorithm demonstrated superior performance.