Dynamic routing optimization based on real time Adaptive Delay Estimation for wireless networks

• Main Authors: Ziane, Saïda, Mellouk, A
• Format: Conference Proceeding
• Language: English
• Subjects: Delay; Delay optimization; Floods; Quality of Service based Routing; Real Time services; Reinforcement Learning
• ISSN: 1530-1346; 2642-7389
• DOI: 10.1109/ISCC.2010.5546584

With the wide emergence of real time applications in mobile ad hoc networks, delay guarantees become increasingly required. Many routing protocols are proposed, in the few last years, for improving the overall delay in mobile ad hoc networks. In this paper, we propose an extension of our earlier QoS routing protocol called "AMDR" (Adaptive Mean Delay Routing) which is based on an adaptive approach using mean delay estimated proactively by each node. AMDR is built around two modules: Delay Estimation Module and Adaptive Routing Module. The first one calculates proactively mean delay at each node without any packets exchange. The second one will then exploit mean delay value. It uses two exploration agents to discover best available routes between a given pair of nodes. Numerical results obtained with NS simulator for different levels of traffic's load show that AMDR improves clearly performances compared to other approaches.