Many practical mobile ad hoc networks (MANET) have certain tasks instead of just randomly changing each node’s positions. We call such a mission-driven network task-adaptive MANET. A typical example is the flying ad hoc network (FANET) that consist of unmanned aerial vehicles (UAVs), which may change its network topology based on different task requirements. Each node moves to new locations based on the targeted network shape. To maintain a smooth topology transformation and minimize the position changes, during shape change, a MANET typically keeps the core-area nodes more stable and allows the nodes in the outer area of the network to move more drastically. This means the entire network has an approximate framework that reflects the relatively stable nodes located in the core area. This research proposes a new routing scheme to quickly identify the optimal end-to-end path using the network framework extraction result. The proposed routing scheme ensures that the packets flow along the more stable network regions (thus with a lower packet loss rate). The framework extraction scheme is based on network shape geometry analysis for the median axis recognition. Our work has contributions to three aspects of realistic network protocol applications: (1) Provides a network multi-center election and member control methodology with detailed protocol design. (2) Creates a stable and reliable MANET framework extraction algorithm which aids in routing table generation. (3) Real-time Unix system protocol implementation and emulation based on Common Open Research Emulator (CORE) + Extendable Mobile Ad-hoc Network Emulator (EMANE). Simulation results indicate that our framework-based routing scheme outperforms a popularly used mobility-adaptive MANET routing scheme—OLSR (optimized link state routing)—in terms of throughput and delay.
Read full abstract