Study On Improved MAC Protocols Of VoWLAN System

With the rapid development of wireless communications technology and mobile terminal technology, the wireless local area network (WLAN) based on IEEE 802.11 is finding more and more applications. At the same time, with low price and high resource utilization, VoIP becomes one of the network applications experiencing tremendous growth. Driven by both technologies, VoWLAN (Voice over WLAN) is also on the rise. Now that IEEE802.11b has become the de facto standard for the MAC of WLAN,and there is demand of supporting voice services on WLAN, however, IEEE802.11b uses shared competitive channel suitable for non-real-time data traffic, and it is difficult to meet QoS requirements of voice,which is sensitive to the delay and delay jitter. Thus, it is important to study voice transmission with quality provisioning in IEEE802.11b WLAN. Our research is carried out on this background.At first,this thesis introduces wireless LAN technologies and the IEEE802.11 standards, with focuses on the IEEE802.11b DCF mechanism. Then, technologies related to VoIP are presented, with details on the key technologies,such as voice coding and signaling access technologies,and a brief introduction on the speech quality evaluation. Based on the above discussions, we estimate the capacity for existing 802.11b, and show that it is inefficient in supporting real-time voice traffic. In light of the characteristics of voice traffic, we propose two improved schemes on the MAC, namely, TD-RAM and TD-TDMA, and analyze their capacity. Qualnet simulator is used to evaluate the proposed schemes, in terms of packet loss rate, the average end-to-end delay, and average jitter performances. The upper bound capacities under different coding standards are obtained, and the theoretical analysis results are verified. The simulation results show that the improved schemes do increase the system capacity. This thesis also states that, in order to increase capacity significantly, improved MAC schemes should be combined with cost-reducing mechanisms, such as increasing voice frame size, compressing the headers and using short PLCP PPDU format in physical layer.