Based On Business Priorities Of Ad Hoc Network Qos Guarantee Strategy And Method Of Research

With the development of wireless network technology, the service provided by Ad hoc network extends from simple data transactions to various service types, such as voice and multimedia. Due to the characteristics of Ad hoc network, such as dynamic network topology, complex estimation of available bandwidth and limited energy of nodes, it keeps to be an important issue to provide effective QoS assurance in such networks. The strategies and methods for QoS guarantee in Ad hoc networks involve all layers of the network, the architecture of QoS model usually corresponds to the five-layer model of TCP/IP. Researches are concentrated at network and data link layer on admission control, QoS routing, rate control, congestion control and packet scheduling.First, we give a QoS management framework for Ad hoc networks. Then, we focus on QoS guarantee strategies and methods based on this framework, including the following three aspects.(1) A traffic priority metric is designed, which combines the importance level of user and the type of traffic.(2) Various congestion control methods for Ad hoc networks are analyzed, and a congestion control method based on bandwidth estimation is proposed and simulated. In this method, congestion is detected by comparing the channel utilization with a threshold. When congestion occurs, we slow down or drop the low-priority traffics to guarantee the QoS requirements of the high-priority traffics.(3) Based on the analysis of existing bandwidth estimation methods for Ad hoc networks, a MAC information based bandwidth estimation method is simulated, which considers the collision probability and back-off time of packets, and is more accurate than those that only consider channel utilization. This method estimates the available bandwidth more accurately, and helps to improve the performance of admission and congestion control modules for different traffics. Finally, we verify the functionality and performance of these QoS guarantee methods by using OPNET simulation platform. The results show that this QoS management framework could distinguish among different importance levels of users and different types of traffics. With the help of congestion control module, the QoS guarantee of high importance level services is more effective in the condition of heavy network load. The bandwidth estimation method, which considers collision probability and back-off time, performs better in distinguishing traffics with different importance level. The combination of various QoS assurance strategies, which combines admission control and packet scheduling, achieves the purpose of ensuring the QoS of high importance level traffics.