Voice Quality Enhancement Technology For Wireless VoIP

VoIP is the technology of using the Internet for voice transmission. The analog audio signal will be compressed and encapsulated into a data packet and then be transmitted in the IP network. Compared with traditional telephone, it's cheaper and can be more convenient to introduce new multimedia services.With the development of the audio/video compression algorithms and high-quality network transmission technology, VoIP applications has gradually matured. Extensive use of wireless networks also makes VoIP more flexible. There is a growing demand to provide real-time voice business in wireless multi-hop networks, but the network congestion, differences of transmission path and other reasons will lead to packet loss and delay spikes. So the voice quality can't be ensured.Voice is a latency-sensitive business, the large delay will affect the interaction and comfort of voice business. In VoIP system, UDP is used to send fixed-rate data stream. Due to the lack of congestion control strategy, network congestion is often happened. When the networks are saturated, the node will continue to inject data packets to the network in an predetermined rate, causing local congestion in some nodes, thus the network delay and packet loss rate will increase rapidly. This phenomenon is called Delay Spike. It will seriously affect the voice quality. This article presents a source rate adjusting strategy to effectively address this issue.In this article, the composition of VoIP systems and the basic principles are discussed firstly, the key technology used in VoIP is described in detail including the signaling technology, the real-time transport protocol, as well as the codecs in the second chapter. The author has made an intensively investigation on the factors affecting VoIP voice quality and voice quality enhancement technology and represented them in the third chapter, including the packet loss concealment/Recovery technology and the Jitter Buffer technology. The fourth chapter explains the delay spike phenomenon in the wireless network and its causes in-depth. On this basis a state-aware and rate-adaptive strategy is proposed. The sending rate of the source voice is dynamically adjusted according to network status in order to effectively alleviate network congestion and to maximize the improvement of voice quality. The strategy can also be used for other real-time business areas, such as video services, multimedia conferencing services and so on.