Predictive QoS for AOMDV protocol based on AutoRegressive Moving Average processes

Mobile Ad hoc Networks (MANET) are wireless networks consisting of a collection of mobile nodes with no fixed infrastructure or centralized administration. With this dynamic topology, ad hoc network presents many specific problems which had influence on solution that assure QoS such as multimedia applications. However, the rapid growth in number and diversity of real-time network applications has made it imperative to consider the impact of end-to-end delay requirements of traffic on network. The main problems are: node mobility and link failure. Hence, there is a need for efficient routing protocols to allow the nodes to communicate with some QoS guarantee. In this work, the AutoRegressive Moving Average (ARMA) model is used to forecasting resources to meet the QoS requirements in ad hoc networks. We apply this model in Ad hoc On demand Multipath Distance Vector (AOMDV) protocol. The results obtained show that the combination of the AOMDV protocol with the time QoS forecasting mechanisms for real-time service support based on ARMA processes performs better than based on AR processes and the conventional AOMDV.